Abdolah CHalehchaleh

امروزه شبكه‌هاي جهاني وب تبديل به يكي از ابزارهاي مورد نياز بشر شده‌اند كه توسط كاربران بسياري در سراسر جهان مورد استفاده قرار مي‌گيرند. مساله‌اي كه در اين حوزه وجود دارد گستردگي بسيار زياد اينترنت و مطالب آن است كه اين گستردگي روز به روز و با سرعت بسيار زياد در حال افزايش است. اكنون يكي از مشكلاتي كه پديد مي‌آيد، اتلاف وقت كاربران براي دستيابي به كالا ها و خدمات مورد نياز آن‌ها است و ممكن است در بسياري مواقع كالا يا خدمت مورد نياز خود را پيدا نكنند، در نتيجه ارائه و پيشنهاد كالا يا خدمات مناسب به كاربران در زمينه هاي مختلف مطابق با نيازها آن‌ها امري بسيار حياتي محسوب مي گردد و يكي از روش هاي بسيار پركاربرد براي اين مساله استفاده از سيستم هاي توصيه‌گر مي‌باشد. ?سيستم‌هاي توصيه‌گر به منظورجلوگيري از اتلاف وقت كاربران، محصولاتي را به آنها پيشنهاد مي‌كنند كه به احتمال زياد مورد علاقه آن‌ها هستند و هنوز آن‌ها را نديده‌اند. الگوريتم پالايش گروهي به عنوان يكي از معروف ترين و پركاربرد ترين الگوريتم‌ها براي پياده سازي يك سيستم توصيه‌گر شناخته مي‌شود. سيستم‌هاي مبتني بر اين الگوريتم بر اساس سوابق جستجو و ابراز علاقه‌مندي كاربر به كالا‌ها و خدمات مختلف و با توجه اطلاعات دريافتي از ديگر كاربران با علاقه‌مندي‌هاي مشابه به كاربر هدف پيشنهادات جديدي را ارائه مي‌كنند. سيستم‌هاي توصيه‌گر اغلب با چالش مهمي به نام تنكي داده‌ها مواجه هستند، زيرا ماتريس‌هاي تعامل كاربر–اقلام در پياده‌سازي‌هاي واقعي معمولاً بيش از ??? تنك هستند. اين مسئله تأثير منفي بر دقت و كارايي توصيه‌ها دارد، به‌ويژه براي كاربران جديد و محصولات خاص يا كم‌تعامل. در حالي كه روش‌هاي موجود تلاش مي‌كنند با افزودن اطلاعات جانبي يا تغيير در طراحي سيستم اثر تنكي داده را كاهش دهند، اغلب مشكل اصلي يعني كمبود داده‌هاي تعاملي را ناديده مي‌گيرند. در اين پژوهش، يك چارچوب جديد معرفي مي‌شود كه از يك مدل ديفيوژني براي توليد امتيازدهي‌هاي مصنوعي با كيفيت بالا در سيستم‌هاي توصيه‌گر استفاده مي‌كند. به طور مشخص، از يك مدل ديفيوژني طراحي‌شده براي داده‌هاي جدولي استفاده مي‌كنيم تا توزيع مشترك و پيچيده‌ي سه‌تايي‌هاي كاربر–اقلام–امتياز را ياد بگيرد و سپس امتيازهاي مصنوعي توليد كند كه از نظر آماري سازگار هستند. براي تضمين كيفيت امتيازهاي توليدشده، يك روش نوآورانه براي شناسايي نويز بر اساس تحليل الگوهاي رفتاري پيشنهاد مي‌كنيم. اين روش امتيازهايي را كه با ترجيحات كاربران و ويژگي‌هاي آيتم‌ها همخواني ندارند، شناسايي كرده و حذف مي‌كند. براي ارزيابي كارايي چارچوب پيشنهادي، از مدل‌هاي پالايش گروهي سنتي و پالايش گروهي مبتني بر شبكه‌هاي عصبي استفاده شده است. آزمايش‌ها روي دو مجموعه‌داده واقعي با سطوح مختلف تنكي داده (با نگهداشت داده از 5% تا ???%) انجام شد و بهبودهاي قابل‌توجهي را نشان داد. به طور خاص، پالايش گروهي سنتي مي‌تواند خطاي RMSE را تا ??% كاهش دهد و نيز پوشش امتيازدهي را تا ?? % افزايش دهد. همچنين، پالايش گروهي عصبي پاسخ‌هاي دقيق‌تر و ظريف‌تري ارائه مي‌دهد و زماني بيشترين كارايي را دارد كه نسبت افزايش داده پايين باشد. اين موضوع نشان مي‌دهد كه مدل‌هاي مختلف به انواع متفاوتي از افزايش داده نياز دارند.   

includes 2605 training samples and 1321 test samples. The labeling of these data was done with two classes positive and negative bythree annotators using the majority voting method. In this study, five different architectures, namely Bert-fa-zwnj-base, Bert-fa-base-uncased, LSTM, GRU, and Distil-bert, were employed for sentiment analysis, and these models were evaluated with two optimizers, SGD and Adam. The results indicate that the Bert-fa-base-uncased model performed the best on both datasets, achieving an accuracy of 93% on the Twitter dataset and 80% on the Instagram dataset. Furthermore, the Adam optimizer outperformed SGD. This research demonstrates that the use of deep learning-based models, especially Bert-fa-base-uncased, can effectively perform sentiment analysis on Persian texts with high accuracy and efficiency, processing data generated on widely used platforms such as Instagram and Twitter effectively.  

Abstract Speech signal processing, as one of the key domains in artificial intelligence and data science, has gained significant importance in today's world. This field plays a crucial role in enhancing human-computer interaction. One of the most challenging and fascinating applications in this domain is speech emotion recognition (SER), which finds applications in psychology, customer service, mental health diagnosis, and even information security. Emotions are intense and specific mental activities that can often be expressed through various behaviors, with speech being one of the most prominent. A speech emotion recognition system works by first receiving a person's voice as input and then determining the emotional state present in the speech, such as anger, fear, happiness, neutrality, etc. In this study, an innovative system based on deep learning methods is proposed, leveraging important and diverse speech signal features such as Mel-frequency cepstral coefficients (MFCCs), Mel spectrograms, zero-crossing rate (ZCR), and others. The system utilizes a dual parallel pipeline for emotion recognition. The first pipeline analyzes the temporal sequence of the speech signal features and sends them to a Bi-LSTM network to model temporal dependencies within the data. The second pipeline performs statistical computations on the extracted features, creating a one-dimensional feature vector, which is then fed into a multilayer perceptron (MLP) model. Finally, the outputs of both pipelines are combined and used for emotion recognition in the speech signal. For evaluation, leave-one-speaker-out (LOSO) cross-validation is employed, and two metrics, WAR and UAR, are used. The results on the EmoDB dataset achieved 83.44% WAR and 80.79% UAR, while on the SAVEE dataset, the results were 58.75% WAR and 54.64% UAR. The findings demonstrate that combining temporal and statistical information from speech signals yields higher accuracy compared to single-method approaches. These results highlight the high potential of this method for practical applications.   

In recent years, people's use of digital products such as websites and mobile applications has increased significantly. On the other hand, the way users interact with the product is one of the important factors in its success. As a result, experts in this field are trying to improve methods and standards related to user experience design by analyzing and measuring criteria. One of the effective methods in user experience research is the use of a questionnaire so that the target users and their needs can be properly met so that the final product has the desired function for the users. There are different questionnaires, each of which has examined user feedback from a specific aspect such as aesthetics, usability, or emotions. However, the modular evaluation questionnaire of the key components of the user experience, or meCUE for short, tests different dimensions effective in the user experience in a simultaneous and standard way. This questionnaire, which has obtained good results in the German language in various tests, was then translated into English and Indonesian languages according to a reliable process in two other articles, and the quality of the questionnaire in the target language was evaluated with criteria such as reliability, Cronbach's alpha test, and measurement. Internal consistency has been evaluated. In this research, three translators first translated the meCUE questionnaire into Farsi based on the international principles of cross-cultural adaptation and evaluated it with various criteria including reliability, Cronbach's alpha test, and internal consistency. To evaluate the questionnaire, it has been used on a comprehensive Iranian application called Rubika with more than 30 million active installations. In this way, 30 application users answered the questionnaire first, and using the obtained results, the criteria mentioned above were calculated to ensure the accuracy of the translation. This questionnaire can be used as the first step in user experience research. In the next step, changes were made in the user interface of the application according to the answers of the users to the questionnaire and also by using the interview tool and Nielson's exploratory evaluation principles. Then the changes applied by 14 people were evaluated with the help of usability testing. To test usability, both people who have not used Rubik's and people who were users were used. In this way, the bias and familiarity of the participants towards the current state of the application are adjusted, the current and potential users of the application are equally considered and a more accurate evaluation of the applied changes is obtained. The results of the usability test showed that the changes applied to the Rubika user interface increased the usability score by 45% on average. In addition, the users who participated in the usability test answered a series of questions regarding the comparison of the current user interface and some proposed changes, as a result of which all the proposed changes were evaluated positively by the majority of users. Keywords: user experience evaluation, human-computer interaction, meCUE questionnaire, user experience improvement of Persian smartphone application, usability testing, Iranian mobile phone application   

رايانش مه يك زيرساخت توزيع شده با امكان ارتباط، ذخيره‌سازي و محاسبه در لبه يك شبكه محلي و بسيار پوياست. فاصله زياد سرويس­گيرنده­هاي يك محيط محلي با ابر و همچنين تعداد بسيار بالاي درخواست‌ها از ساير محيط­ها كه حساس به تأخير هستند مشكلاتي را در ارائه خدمات ابري به وجود آورده است. درنتيجه استفاده از قابليت محاسباتي منابع بيكار محلي و نزديك به دستگاه‌هاي انتهايي همانند خودروهاي با/بدون سرنشين و ايجاد يك شبكه ad-hoc تحت عنوان رايانش مه وسايل نقليه سبب كاهش ارسال درخواست‌ها به ابر و همچنين كاهش زمان پاسخ مي‌شوند. با اين ‌وجود محدوديت منابع در رايانش مه وسايل نقليه در مقايسه با ابر باعث ايجاد مشكلاتي از قبيل يافتن منابع آزاد از نظر توان محاسباتي و همچنين دسترس‌‌پذيري منابع در ارائه سرويس مطلوب به مشتري­ها مي­شود. درنتيجه تلاش براي پيش­بيني درست ميزان منابع درخواستي هر وظيفه مي­تواند از هدر رفتن منابع محدود گره­هاي مه جلوگيري كند كه اين امر نيازمند روش‌هايي از قبيل يادگيري ماشين است تا بر اساس درخواست/پاسخ‌هاي دستگاه‌هاي انتهايي بتواند رفتار محيط را تا حدودي ياد گرفته و جهت رسيدن به كيفيت مطلوب سرويس­دهي، مقادير مناسبي از منابع را در اختيار آن­ها قرار دهد. در اين پژوهش با استفاده از يادگيري تقويتي عميق QL روشي براي برنامه‌ريزي و پيش‌بيني منابع مورد‌نياز يك مشتري خودرو هوشمند با معماري سه‌لايه رايانش خودرويي در بهينه‌سازي تخصيص منابع و بهبود عملكرد كلي سيستم ارائه شده است.اين روش با استفاده از قابليت‌هاي هوش مصنوعي و يادگيري تقويتي، رويكردي پويا و تطبيقي براي مديريت منابع در يك محيط محاسباتي مه ارائه مي‌دهد. دو الگوريتم اصلي براي مسئله پيش‌بيني و تخصيص منابع در اين تحقيق پيشنهاد شده است. در انتها بر مبناي روش پيشنهادي از ابزارها و مجموعه داده­هاي مناسب جهت ارزيابي استفاده مي‌شود. داده‌هاي مورداستفاده هم مي‌توانند يك بازه‌اي مشاهده شده از دنياي واقعي باشند و هم مي‌توانند از طريق ابزارهاي شبيه‌سازي مانند Matlab توليد شوند. ديتاست مورداستفاده شامل وضعيت‌هاي خودروهاي كلاينت، درخواست‌هاي آنها، گره‌هاي مه، تحركات آنها و وظايف درخواست شده از سمت كلاينت‌ها در يك بازه زماني خاص است. يافته‌هاي كليدي اين مطالعه نشان مي‌دهد كه يادگيري تقويتي QL مي‌تواند به طور مؤثري ميزان متناسب تخصيص منابع را با يادگيري از تجربيات گذشته و تصميم‌گيري آگاهانه پيش‌بيني كند. با آموزش و به روزرساني مستمر عامل يادگيري Q، سيستم مي‌تواند با شرايط متغير سازگار شود و تصميمات تخصيص منابع را بر اساس اطلاعات بلادرنگ اتخاذ كند. همچنين نتايج آزمايش‌ها اثربخشي روش پيشنهادي را در بهينه‌سازي تخصيص منابع نشان مي‌دهد. عامل يادگيري تقويتي QL اقدامات بهينه‌اي را ياد مي‌گيرد كه مصرف منابع را به حداقل مي‌رساند درحالي‌كه الزامات عملكرد سيستم مه را برآورده مي‌كند. اين منجر به بهبود كارايي، كاهش تأخير و افزايش قابليت اطمينان سيستم مي‌شود.   

   The importance and increasing energy crisis in the current world is increasing, so various methods have been used to reduce and optimize energy consumption. In the meantime, the importance of the construction industry, especially office buildings, in energy consumption is very significant and also has the largest share in the production of greenhouse gases. This indicates how important research is in the field of optimal building design and saving or reducing energy consumption during operation. So far, many methods and solutions have been presented to reduce and optimize energy consumption through the principled design of buildings. The flexibility and versatility of the interior spaces of buildings are among the solutions that have not been sufficiently considered while being effective. Office buildings are among the buildings that have a variable number of users at different times and according to the change in conditions at different times, and therefore the number of people using them cannot be considered a specific number when designing these buildings. The flexible design of the interior spaces of office buildings can be effective in responding to the change in the number of people at any given time. But the important issue in using this solution is that a number or parts of interior spaces are not used at different times and at the same time these spaces benefit from installation systems along with other spaces. However, such spaces can be separated from other spaces by movable internal walls and removed from the cycle of using installation systems, and as a result, less energy can be used to meet the heating and cooling needs of controlled spaces.This research has been formed with library methods and the use of simulator software with the aim of discussing the effect of flexibility of interior spaces of office buildings on energy consumption optimization and discusses this issue through the necessary simulations with the relevant software.

Abstract Objective: Due to the increasing growth of image transmission in computer networks, it is very important to provide a suitable level of security to protect these images, which can be ensured by using different encryption methods. Image encryption methods based on chaos theory are known as a more effective and safer solution in image encryption due to the unique characteristics of chaos functions, such as sensitivity to initial values and parameters and high scrambling power. This study was conducted with the aim of determining the optimal chaotic mapping for encryption of four different groups of images including face images, fingerprint images, satellite images and medical images and increasing the encryption speed. Research method: First, texts and articles related to image encryption were studied using chaotic mappings. Using these studies, 11 one-dimensional and two-dimensional chaotic maps were investigated. In the implementation phase, 40 images were encrypted with these 11 maps using the Python programming language. Then, in the evaluation stage of encrypted images, the encryption quality was checked with the help of criteria such as image histogram and correlation between image pixels. After the evaluation stage, it was determined that for the encryption of each of these image grou   which mapping is more appropriate? In the end, the encryption speed was increased by using parallelization techniques. Findings: The result of this study was to determine the appropriate chaotic mapping for encryption of each of the four image groups and the parallelization of chaotic key generation. Also, the chaotic function of sinusoidal mapping was improved by making changes in the equation of this mapping. After analyzing the encrypted images, it was determined that Logistic 2, Logistic 3, Duffing, and Sinusoidal mappings are the optimal mappings for face, medical, fingerprint, and satellite image encryption, respectively. It was also found that chaotic quadratic mapping has the highest speed of generating chaotic keys. Conclusion: One of the available methods to ensure the security of images during transmission in computer networks is image encryption. In image encryption, one of the important steps is to generate encryption keys. Pseudo-random keys can be generated by using chaotic functions. There are different types of chaotic functions that it is better to choose a suitable function for image encryption according to the type of image. The use of chaos functions increases the security factor of image encryption; but it usually requires a lot of calculations. This volume of calculations can reduce the encryption speed. To solve this problem, different parallelization methods can be used. Keywords: image encryption, chaos functions, optimal mapping, parallelization   

توسعه زندگي بشري منجر به ايجاد رخدادهايي متنوع در سطح جامعه شده است، دولت ها جهت كنترل اين رخدادها موجودتي به نام قانون را ايجادكرده اند تا به وسيله آن، رخدادهاي بشري را كنترل كنند. از اين حيث شناخت دقيق قوانين جهت دفاع از حقوق فردي، جمعي و يا قضاوت، رخداد ها بر اساس اين قوانين امري بسيار پيچيده است. چرا كه استنباط هر شخص از رخداد و قوانين بر اساس دانش، تجربه، شخصيت و احساسات است. با افزايش اين رخدادها خصوصا رخدادهاي يكسان و به طبع آن افزايش پرونده هاي دادرسي، شواهد و نظرات متنوع نسبت به رخدادها، منجر به ذهني شدن رسيدگي به رخدادهاي يكسان شده است، از اين رو بنا بر اينكه عدالت در صدور آراي قضايي مهمترين اولويت يك دستگاه قضايي است ذهني شدن قضاوت در پرونده هاي مشابه، عدالت در صدور آراي قاضيي در پرونده هاي مشابه را زير تحت تأثير قرار ميدهد. وجود ابزار و الگوريتم هاي شباهت سنجي با استفاده از هوش مصنوعي ميتواند جهت استفاده كارشناسان حقوقي و نيز دادخواهان بسيار مفيد واقع شود.   اين شباهت سنجي به طرفين دعوي، وكلا و قضات كمك ميكند كه آراي صادره نسبت به يك رخداد يكسان را مشاهده كرده و نسبت به آن وحدت رويه داشته باشند. وحدت رويه موضوعي است كه باعث ميشود قضات در تصميم گيري نسبت به پرونده هاي مشابه بتوانند اعمال نظري دقيق تري انجام دهند و در تصميم گيري نسبت به يك موضوع اجماع نظر داشته و در برخورد با موارد مشابه سليقه اي برخورد نشود. در شباهت سنجي قوانين و آراي صاده مشكلات و چالش هاي فراواني   وجود دارد كه يكي از مهمترين آنان عبارت است از زبان قوانين و عدم دسته بندي هاي لازم در اين متنون است. براي ارتباط و شباهت سنجي متون قضايي با وجود محدوديت ها و چالش هاي موجود از يادگيري عميق در زمينه پردازش زبان طبيعي(NLP) استفاده خواهيم كرد. براي پردازش زبان آراي صادره نيازمند به يك الگوريتم پردازش زبان، براي زبان مورد نظر هستيم. استفاده از يك سيستم شباهت سنجي مبتني بر هوش مصنوعي ميتواند به عنوان يك ابزار قابل اتكا براي كارشناسان قضاييي مورد استفاده قرارگيرد.   

Nowadays, social networks have become an inseparable part of people's lives; As far as most people use these networks to express their feelings about all aspects of life. Because understanding and analyzing these emotions has many applications, such as customer relationship management, checking and observing people's mental health, identifying public emotions caused by a national, global or political event, identifying criminals and improving the efficiency of responsive robots, etc.; Identifying the intensity of emotion from the texts of social network users is considered a very practical and important issue. In this project, we are trying to find the most accurate model for predicting the intensity of emotion from social network tweets with the help of different natural language processing methods, using regression-based machine learning models and neural networks. In this research, we first used basic methods such as Bag of Words, Word2Vec, GloVe and TF-IDF and obtained accuracy on the data of the SemEval2018 Task1 EI-Reg competition, then using modern methods such as GPT2 and model Based on BERT, we perform various tests on these data in order to reach the best possible result in terms of Pearson correlation. The result obtained from the methods used, which is a combination method of different models, is equal to 0.82, which compared to the previous works in this research and the teams participating in SemEval competition is the best result.

Accurate and timely diagnosis of brain tumors is essential for effective treatment of this disease. The choice of treatment method depends on the level of the tumor at the time of diagnosis, the type of pathology and its grade. Glioma brain tumor is the most common type of primary brain tumor and all of them originate from glial cells that surround neurons. In diagnosing this type of disease, computer-aided diagnosis methods have helped neurologists in various ways. Recent works in this field have led to improved efficiency with the emergence of the concept of deep learning. Computer-aided recognition systems approaches include preprocessing, segmentation, feature analysis (feature extraction, feature selection, and feature verification) and >   In this thesis, methods based on image processing, machine learning, and deep learning have been used to identify glioma brain tumors and >The dataset used in this research is Brats2018, which includes 210 MRI images of high-grade glioma tumors and 75 images of low-grade glioma tumors. In the first proposed method, image segmentation was done using the combination of K-Means clustering algorithm and Coyote optimization algorithm, as well as different feature extraction methods, including texture feature extraction using local binary patterns method and deep feature extraction. From the trained neural network and the pre-trained VGG16 network, among the mentioned methods, the extraction of deep features resulting from the precise adjustment of the pre-trained VGG16 network brought the best results. This proposed method reached an acceptable accuracy of 99%. In the second proposed method, the clustering centers were optimized using the coronavirus algorithm, and the previous feature extraction methods were also implemented for this method, and the best performance related to feature extraction was through a trained neural network. be In this process, we reached 99.80% accuracy. Keywords: Glioma Brain Tumors, Clustering, Coyote Meta-heuristic Algorithm, Coronavirus meta-heuristic algorithm   

پيشرفت روزافزون فنّاوري در عرصه‌هاي مختلف علوم و تأثير آن بر زندگي انسان امروزي، تجارب احساسي، عاطفي و ادراكي را به‌شدت در كانون توجه طراحان قرار داده است. در اين خصوص، طراحي بر اساس رضايتمندي، خوشايندي، احساسات و عواطف دروني انسان عاملي بسيار مهم و تأثيرگذار در فرايند طراحي محصول شناخته مي‌شود. به دنبال شيوع و فراگيري ويروس كرونا در جهان، ساختار آموزش عالي نيز، مانند بسياري از بخش‌هاي ديگر زندگي انسان، دست‌خوش تغييرات عمده شد.   شركت دانشجويان در كلاس‌ها ي آنلاين، آزمون‌ها و انجام امور اداري به‌صورت غيرحضوري موجب استفاده بيشتر دانشجويان از وبسايت دانشگاه‌ها شده است. استاندارد نبودن طراحي وبسايت باعث مي‌شود زمان زيادي از دانشجويان گرفته شود تا به اهداف موردنظرشان برسند. بنابراين گنجاندن عناصر احساسي كه مي‌توانند شادي، لذت و علاقه را تشويق كنند، بسيار مهم است. اين تحقيق از مهندسي كانسي استفاده كرده است تا احساسات كاربر را به مولفه‌هاي طراحي رابط تبديل كند و نشان دهد كاربر از رابط كاربري چه مي‌خواهد. 50 كلمه‌ي كانسي از طريق پرسشنامه بين 50 دانشجو توزيع گرديد و از بين آن‌ها 12 كلمه جهت ارزيابي پارامترهاي طراحي بر اساس احساسات كاربران انتخاب شد. بر اساس كلمات كانسي انتخاب شده پارامترها و قوانيني براي طراحي رابط كاربري استخراج شد. اين قوانين در يك پايگاه دانش جمع‌آوري گرديد كه طراحان مي‌توانند با مراجعه به آن بر اساس احساس موردنظرشان براي طراحي، پارامترهاي طراحي متناسب با آن احساس را دريافت كرده و طرح كاربرپسند خود را ترسيم كنند.   

In recent years, with the growth ofscience and technology and the creation of competitive markets in variousindustries, the need for quality control, measuring the quantitative andqualitative parameters of the final product has become very important. Having aquality product is the most important part of a production line, so that todaythere are few advanced factories where part of the production is not controlledby intelligent machine vision and image processing applications. Qualitymanagement in real time and on line provides the possibility of increasingproduction efficiency effectively.In this thesis, anattempt has been made to research and examine advanced techniques in the fieldsof image processing, machines and learning in order to improve the quality ofwood defect detection as a basic material in the wood products industry. The maingoal of this project is to improve the accuracy and ability to detect wooddefects through the use of advanced tools and techniques in the fields of imageand machine processing. In this study, the "Wood_patches" dataset isused, which includes many images of healthy and unhealthy wood of differenttypes. Also, in order to further evaluate and deepen the effectiveness of themodels, the "Leather Defect" dataset is also used, where there arehealthy and unhealthy leather >In the first proposed approach for predicting wood defects,three main steps are performed for independent feature extraction and>The second proposed approach is prediction by extractingcombined features and >

   Abstract: Today, with the increasing expansion of social networks, users have access to the opinions and views of other people. These opinions often contain valuable information that can be analyzed to understand people's tastes and tendencies and to identify their positive, negative or even neutral opinions on various issues. But since the volume of these data and the speed of their production is surprisingly high, analyzing it by humans is a difficult, time-consuming and practically impossible task; Therefore, there is a need for a system that can automatically analyze comments. Sentiment analysis is the solution to this problem. Sentiment analysis is a process that is able to discover people's views, attitudes and feelings from their writings. Sentiment analysis or opinion analysis is a subset of text mining and natural language processing, the purpose of which is to automatically extract users' views on various issues. Microblog is a type of social network where users try to share their short texts with others. Twitter is one of the most popular microblogs in which the maximum size of each tweet is 280 characters, and this feature has made Twitter a suitable platform for knowing the opinions of users. In this thesis, sentiment analysis has been done on 7306 Farsi tweets extracted from the Twitter social network on the topic of Corona vaccine. For this purpose, tweets were considered in three >Keywords: text mining, sentiment analysis, corona vaccine, natural language processing, machine learning, deep learning, Twitter social network

   Normal 0 false false false EN-US X-NONE AR-SA

Abstract: The ever-increasing population growth in megacities has changed human lifestyle and the need to create high-rise buildings for optimal use of urban land has increased. Therefore, many deep and semi-deep excavations are carried out in these cities. In some areas, the presence of surface and subsurface water in the area of excavation operations causes issues and problems for the implementation of the project, each of which requires the investigation of all aspects affecting the stability and deformation of the pit. The most important forces that cause the instability of pit walls in urban areas with high groundwater levels are (1) gravitational force (2) water leakage force. The presence of water flow around the pit restrained by the nailing system can cause destruction of the pit or damage to nearby buildings. Therefore, one of the practical and economic solutions to create stability in the restrained pit wall is to lower the underground water level to the extent of the excavation stages. This is despite the fact that each excavation operation and reduction of the underground water level alone causes a change in the state of tensions in the soil and can lead to large deformations in the wall and subsidence of the ground around the pit. In this research, with Plexis, Midas and Geostudio software, the step-by-step excavation operation of a pit under the influence of underground water was modeled, and by using the finite element analysis of the stresses in the soil, the deformation of the wall, the bottom of the pit, and the soil affected by the excavation operation. And we have calculated and investigated the forces entering the Niles. According to the findings of the research, the higher the underground water level, the changes in the walls, bottom of the pit and the surrounding ground, the force acting on the Niles increases, the safety factor of the stability of the pit wall decreases. The output of Plexis software differs by up to 50% from Geostudio and Midas, but the process of groundwater impact on pit stability and deformations is the same. Using the hardening behavior and the small strain hardening behavior have almost the same output. The amount of output in Good simulation models with Mohr-Columb behavior pattern is 50% of the output of the same model with hardenable and small strain hardening behavior patterns. And the previous findings regarding the weakness of the Mohr-Columb behavior model were also proven in this research.   

Abstract The term cryptocurrency is an emerging topic in today's world, which has created a revolution in our vision in the field of investment and has caused changes in the world's financial systems. Cryptocurrency is a digital currency that uses blockchain technology with secure encryption. Every change can have advantages and disadvantages, cryptocurrencies are no exception to this rule, and along with their advantages, they can also have disadvantages for the economy of any society, so that due to the decentralization of these currencies, traditional monetary systems and the capital market of each They can influence a society. Therefore, due to the importance of the issue, the need to understand public opinion and analyze people's opinions in this regard increases. To understand the opinions and views of people about different topics, you can take help from social networks because they are a rich source of opinions. The Twitter social network is one of the main platforms where users discuss various topics, therefore, in the shortest time and with the lowest cost, the opinion of the community can be measured on this social network. Twitter Sentiment Analysis (TSA) is a field that analyzes the sentiment expressed in tweets. Considering that most of TSA's research efforts on cryptocurrencies are focused on English language, the purpose of this research is to investigate the opinions of Iranian users on the Twitter social network about cryptocurrencies and provide the best model for >  

Network Functions Virtualization of architecture means providing various network services without the need for hardware and not depending on it. Network Functions Virtualization is a new field in the network, with the help of which hardware devices can be implemented in virtual and software form. Network Functions Virtualization improves network functions such as: proxies, firewalls, load balancing, etc. In other words, using virtualization technology, this architecture is able to convert hardware devices into software modules known as virtual network functions and provide the desired service to the user. Providing the service requested by the user in the network is done by a sequence of virtual network functions, which are known as service functions chain. One of the main challenges in the development of network functions virtualization architecture is the allocation of resources to the requested network services in network infrastructures based on network functions virtualization, this challenge is called network function virtualization resource allocation problem. Therefore, in this research, the problem of allocating resources to virtual network functions in Network Functions Virtualization architecture has been solved by using mathematical programming techniques. In this research, a multi-objective mixed integer linear programming model is presented for the problem of resource allocation to virtual network functions. In this model, constraints related to the resource capacity of nodes and connections and delay constraints are desired. Also, the objective functions in this research are: maximum flows accepted in the network, reduction of resource costs of nodes (including: the number of CPU cores and the amount of memory), reduction capital costs, reduction operational costs and checking execution time. These constraints and objective functions are expressed precisely and explicitly by mathematical functions. The proposed mathematical model is implemented and solved with the Cplex solver. To evaluate the proposed mathematical model, several different topology are considered. The optimal cost is evaluated under changing parameters such as the length of service functions chain, the number of flows, the length of flows, the amount of resources of nodes, the number of nodes and the number of virtual network functions. And finally, the increase in execution time is checked by changing the number of nodes and the number of virtual network functions. The numerical results of this research show the effectiveness of the model in resources allocation to virtual network functions. Keywords: Network Functions Virtualization architecture, Virtualized Network Functions, Resource allocation, Mathematical programming, Mixed integer linear programming   

  Considering the growing trend of the population of the cities and the need for new urban elements and space suitable for the daily needs of the society, having places to establish social and cultural interactions is felt more than ever. In order to achieve this goal of having public spaces, including cultural centers, which lead to environmental quality, it seems essential that different people can interact and spend leisure time with ease.In the city of Kermanshah, due to its historical and cultural background and native and local values, the lack of suitable cultural and artistic centers has caused the cultural identity and cultural and artistic activities of the citizens to fade; On the other hand, there are buildings and spaces with a cultural and historical background and with great potential for development, which are facing wear and tear due to the neglect of the relevant authorities.In order to improve the environmental and cultural qualities of the municipal garden, the present research investigated culture and cultural factors by using the principles of regeneration concept and looking at it in the form of applying culture-oriented regeneration strategies, including the creation of cultural centers and building restoration. which contain part of the history and personality of the city, presents the cultural regeneration plan of the municipal garden and turns it into one of the most successful centers in order to revive the culture, history and art of the city and improve the cultural status of the neighborhoods. It transforms around it.The current research method is descriptive-analytical and the method of collecting information is based on the examination of case samples, field observations, library studies, and the examination of documents, maps, and historical images. The main results that are expected to be extracted from this research are the revival of the cultural and historical identity of the municipal garden and its surrounding neighborhoods and the response to the cultural and social needs of the society, which is achieved through the design of a cultural center with an attractive and dynamic environment and also with benefits. Taking the city's specific cultural and artistic resources and changing the use of the municipal building (former British consulate) into a museum and highlighting this area as one of the strong cultural poles in the city is realized.

   Abstract Today, the expansion of the use of the Internet has made it possible for millions of users around the world to access online social networks. On the other hand, these networks have been in the focus of users' attention in the dissemination of information, especially in areas such as viral marketing advertising, improving recommender systems, transferring time-sensitive information, guiding public opinion, promoting national security, sociology, etc. One of the important issues surrounding the dissemination of information in online social networks is the issue of effective dissemination of the message at a suitable speed. For this purpose, it is necessary to identify the influential users in a suitable way. In this thesis, it is proposed to identify influential people based on the reactive information of users and according to their sphere of influence. For this purpose, first, the value of network communication is determined based on the reactive information of users (reply and retweet), and then, the structure of the network is divided into its constituent communities. In the next step, centrality criteria are used to evaluate the importance of each member of the community, and at the end, influential nodes are identified in their sphere of influence. The effectiveness of the proposed method using a real database that includes the retweet and reply information of Twitter users; has been tested. Due to the lack of a database that has the tagging of influential people and reactive information at the same time; from one of the methods of information dissemination in social networks that has an acceptable correlation with the proposed method; Used. At the end, the obtained results are compared with previous similar methods. The results of the evaluations showed that the method proposed in this thesis can identify effective users with more accuracy, efficiency and speed. Also, the evaluation results showed that the proposed method can achieve 87.33% detection accuracy in identifying effective users, which shows an improvement of at least 13% compared to the compared methods.    Keywords: social network analysis, identification of influential users, identification of communities, graph analysis.   

  Long-ShortTerm Memory neural networks have high computational complexity, resulting inlong execution times. Hardwareimplementation is one of the proposed solutions to this challenge. However, thepower, delay, and area are serious challenges in the hardware design process. Asuitable solution would be to replace the approximate circuits with exactcircuits. The purpose of this study is to evaluate the efficiency of long-shortterm memory networks using approximate computing units. However, because themultiplication operator is used extensively in the network structure, we turnedour attention to replacing the multipliers. For this purpose, we replaced allthe multipliers in the EvoApprox8b approximate library instead of the exactmultiplier in the network structure for forecasting stock market signals in thedatasets of Apple, Microsoft, and IBM companies, and we examined theperformance of the network. From the simulation results, it was found thatreplacing the approximate multiplier can cause a decrease of 205.8 µw in power,0.64 ns in delay, 236 µm2 in area, and 366.5 J in PDP, to predictthe close signal in the Apple stock market dataset. The substitution effect inpredicting the High signal in the Microsoft stock market dataset was in theform of a decrease of 170 µm2 in area, 38.8 µw in power and anincrease of 17.7 J in PDP, and the substitution effect in predicting the Lowsignal for IBM was in the form of a decrease of 47.1 µw in power, 432.646 J wasobserved in PDP and 87 µm2 in area. In the second part of study,with the help of the inherent hardware criteria of the approximate multipliers,we designed a predictive model to predict the suitability of the multiplier forreplacement in the network hardware structure. Therefore, a binaryclassification problem was defined. Next, using feature-selection algorithms,we determined the number of entries in the desired model. Our proposed model isa Linear Discriminant classifier, which can predict the performance of anapproximate multiplier from the EvoApprox8b library in the LSTM networkhardware structure using the Mean_AED, Correct, and Var_ED inherent errorcriteria with 99% accuracy.

The world around us is full of images. Pictures are documents that, by recording a moment, become the narrator of a world of words. City cameras create, record, and store thousands of traffic images every second. Proper processing of these images can help train models based on deep learning. Such models are used in object recognition and image captioning and will be used in cases such as voice assistants and self-driving cars. In this thesis, a method is introduced to convert traffic images into their descriptions. The presented description is based on prominent objects from images and deep learning and includes three basic steps. In the first stage, data processing and methods such as data augmentation are performed on training images. In the second step, appropriate features are extracted from the images. For this purpose, four deep neural networks named VGGNet, EfficientNetB0, InceptionV3, and ResNet50 have been investigated to extract image features. According to the number of layers in the architecture of each of these deep neural networks, the fine-tuning technique has been applied to improve the accuracy of detecting traffic objects. In the third step, two neural networks, LSTM and Transformer, have been used to convert image features into text. Finally, the optimal solution will be introduced, which will significantly increase the quality of the output sentences. In total, two methods were introduced. Based on the Transformer network, the second method showed better accuracy than the first. The MS-COCO dataset was used to evaluate the proposed methods. For this purpose, a subset including 8,000 images and ten classes of traffic objects in the MS-COCO dataset has been separated and pre-processed. The accuracy of the model introduced in the BLEU evaluation criteria is 65.3595%.

   Individual differences originate from one's personality, which is the most crucial factor affecting one's decisions and choices in life. In recent years, automatic personality detection from text has attracted the attention of most researchers due to its applications like recognition of qualified managers, job selection, selection of academic courses, and online businesses. However, most current methodologies focus on the statistical features of text and ignore the semantic relations and information. These writings and texts are a way to translate internal thinking and feeling in a manner that is understandable to others. The research aims to automatically predict personality from text using language modeling and deep learning. This research presents a deep learning algorithm, Convolutional Neural Networks, through two approaches for personality prediction from two benchmark datasets: Essay and MyPersonality. In the first approach, whole text is utilized for modeling, training and evaluation; However, in the second approach, key phrases derived by a multipartite graph are used. In order to extract features, three techniques of SentenceTransformer, Longformer and short-time Fourier Transform are presented and applied for the first time in the personality prediction research.

AbstractThe aim of non-parallel voice conversion (VC) is to train a voice convertor without relying on paralleldata. Due to the good performance of MaskCycleGAN which is a family of CycleGAN-vc, we usedit as our baseline system here. MaskCycleGAN was an improvement of CycleGAN-vc2, by replacingmel-cepstral features with mel-spectrogram ones, benefited from a mechanism, “filling the frames”(FIF) that make the convertor fill the artificially made missing frames based on neighboring frames.However, this model was not able to capture the inter-channel dependencies. To do so, we proposean attention mechanism integrated in the convertor to help it enhance the sensitivity of the networkto more important features. This application of attention mechanism has a twofold advantage. First,there is no need to define new models, subsequently a vast number of parameters will not be imposedon the network, and second, the process of updating the parameters is done through back propagation.A subjective evaluation of the similarity and naturalness, as well as an objective evaluation shows thatour proposed model outperforms the conditional MaskCycleGAN.

  With the size of transistors reaching the nanometer scale, a large number of processing units (PEs) can be embedded in one chip. With the increase in the number of PE in a chip, the existence of a scalable and powerful communication infrastructure for communication between these cores is necessary. Network-on-chip (NoC) is proposed to meet this need. However, like any other electronic component, the network-on-chip is prone to transient and permanent faults. Among the transient and permanent faults, the transient faults have a higher rate. These faults can reduce the performance of the network-on-chip, and if the rate of faults increases, the whole system is susceptible to failure. Most of the past solutions in the field of fault-tolerance network-on-chip, which have worked in the field of router design, have investigated the problem of single-bit faults. This is despite the fact that due to the shrinking size of transistors, the probability of multi-bit transient faults has increased. Therefore, in this thesis, we have used the Flexible Unequal Error Control (FUEC) coding method in our proposed router to correct single-bit, two-bit, and three-bit faults in the channel and the router's input buffers. We have also used the Triple Module Redundancy (TMR) technique to deal with the problems of router control units. Also, to correct the faults on the multiplexers of cro  ar, we have used redundant information and redundant time to discover and correct the faults, respectively. However, all fault tolerance techniques are associated with hardware overheads. These overheads increase power consumption, while power consumption is an important challenge in today's world. Therefore, it is very necessary to have mechanisms that are power-aware to reduce power consumption. Therefore, in this thesis, a power-aware mechanism has been proposed, which reduces the energy consumption by turning off the idle input buffers in the routers. To evaluate the proposed router, we have implemented it in the Noxim simulator and in this simulator, we have randomly injected faults for synthetic traffic and 8x8 two-dimensional mesh. On the other hand, using the SystemVerilog language, we have synthesized the hardware design of the router in the Vivado2019.1 and obtained the hardware overheads and power consumption using this tool. Using this router in the on-chip network can increase fault tolerance by 5 times and reduce energy consumption by 12% compared to the basic router. On the other hand, this router requires 57% more hardware overhead than the basic router. It also increases the average network delay up to 2 times.

Abstract

  The amount of text that is produced every day increases dramatically. Therefore, efficient and effective techniques and algorithms are needed to discover useful patterns. With the pervasiveness of social networks in recent years, despite their positive applications, spreading rumors has become easier and more common. Rumors are a security challenge on social media, as a malicious node can easily discredit or isolate its goals by spreading a rumor. Therefore, detecting rumors is an important challenge in soft security mechanisms such as trust and reputation. In this study, we used the machine learning approach of the LSTM deep learning model and deep neural network to simplify feature extraction and create a strong ability to learn, and automatically detect features compared to traditional machine learning methods. LSTM neural network Due to its special architecture, it is very suitable for working with sequential data, especially textual data. But the performance of this network is highly dependent on the regulation of its hyperparameters, so a new approach to improve the result using a genetic algorithm to regulate the hyperparameters of the deep neural network is proposed. The standard genetic algorithm has its own problems, including the speed of convergence in this algorithm. In this study, we have solved this problem by adjusting and formulating the rate of algorithm processes based on two criteria of suitability and diversity, and we have reached a detection accuracy of 0.93%. ام.

Pilates includes set of motions, Focusing on the simultaneous useof mind and body, to increase the body's resistance,. uses gravity, body weight andspecial devices. If for any reason a person intends to perform these movementsat home without a trainer, there are various commercial software that play therole of trainer; But these softwares do not have a guide or software monitor togive the user proper feedback on the correctness of the movements. Thisdissertation addresses the issue of the correctness of the user's Pilatesmovements by providing an approach based on image processing techniques.In this study, computer analysis of 6 main Pilates movements in theabsence of a Pilates instructor has been performed. To do this research, asuitable data set is needed first. To collect data, 20 main body joints wereextracted from deep video in each 3D frame using a Kinect sensor. The resultingdata set contains 300 records that are collected in a fixed location fromdifferent users. The proposed method consists of four steps. First, thethree-dimensional coordinates of the 20 main joints are extracted from theinput. In the second step, the required preprocessors include calculating the 4main body angles, namely the angles of the knees and elbows in each frame,applying the Savitzky Golay filter, and the PiecewiseAggregate Approximation. In the third step, various functions were proposed to calculatethe data distance, which are Dynamic TimeWarping, Hausdorff,fast Dynamic Time Warping, and providean improved distance function based on fast Dynamic Time Warping. In the fourth stage, learning and >After >  

nowadayswe have entered a new era of information exchange due to the widespread use of mobile devices andthe Android operating system is the most popularmobile operating system in the world. Simultaneously with applications, manymalicious applications with different purposes and forms for the Androidoperating system are being developed and released. Despite the increasing developmentof Iranian programs in software stores, it has never been investigated how muchof the malware is possible among them, that it may endanger the safety of usersor with other targets such as high volume of advertising, to offend users. For this reason, we decided tocreate models based on the static permission feature using a secure data setusing nine machine learning classifiers as well as a deep learning approach toclassify more than four hundred Iranian applications which wererandomly downloaded from the Cafe Bazaar store, in two categories: maliciousand Benign application. Then we will analyze the results and also by making allthe mentioned models on the samples downloaded from the Cafe Bazaar store, wewill complete our evaluations on the effect of the permission feature indetecting Iranian malware. Theresults obtained from the models made as well as the scanned samples downloadedfrom Cafe Bazaar store in the reputable site of Virus Total show that more thanfifty percent of the samples downloaded from Cafe Bazaar store are malware. Therefore, in order to increase theconfidence of Iranian users that the downloaded Android applications are notmalicious, the current approach to screening applications should bereconsidered before placing them in software stores.

   In today's information age, it is a prerequisite that we have reliable information before making any decision. In this regard, location-based social networks have become an important program in location-based social networks as an effective way to help users find attractive places and recommend points of interest. Recently, they have gained a lot of popularity. Adding a location dimension to these networks makes their information closer to reality by creating a bridge between virtual social networks and the real world. The purpose of creating these networks is to provide location-related services; By allowing users to share experiences and visited locations with other users in different geographical locations. Location-based social networks are rich resources for data mining and information discovery by obtaining and updating the information of their users around the world. Recommended systems are also a special type of intelligent systems that take advantage of users' past rankings. Collaborative filtering is one of the most common approaches used for recommendation systems, although this method can sometimes present challenges such as cold start. Cold start occurs due to data scatter and is based on the fact that most users only connect to a small number of possible locations and the recommendation system for ranking some items or new users lacks data; Not available or only a small amount of data available. Solving this problem can greatly improve the user experience and trust in recommender systems. In this dissertation, we try to use machine learning and deep learning algorithms to provide a spatial recommendation system with a participatory filtering approach. Therefore, by implementing the torsional neural network algorithm on Yelp data set and presenting experimental results, we show that the proposed method can perform better than other related methods. Keywords: recommendation system, collaborative filtering, location recommendation, location-based social networks, deep learning, convolutional neural network

  Abstract Images are the most important and widely used digital data used in computer systems. A digital image is made up of a set of objects or areas, so one of the efficient techniques for extracting features from images with respect to their constituent objects is the image segmentation technique, which delimits objects or areas. Highlights the image with high accuracy due to its texture and features. Using image segmentation, image pixels are placed next to each other in specific areas due to common features and generally similarity to each other. Multi-level image thresholding is one of the most popular and at the same time the simplest and most efficient methods of image segmentation. The most important issue in this method is the selection of the value of the relevant thresholds. In such a way that by determining the appropriate thresholds, the desired image can be more accurately zoned. Atsu method is one of the thresholding methods that has a good performance in determining two-level thresholds, but when increasing the number of thresholds, Atsu performance decreases in terms of time and segmentation accuracy. Therefore, it is combined with optimization algorithms to achieve better performance in terms of time and segmentation accuracy. In this research, an improved Grasshopper optimization algorithm is also proposed to increase the accuracy of finding answers and increasing the accuracy of segmentation, as well as to increase image quality. In this method, Atsu evaluation function is proposed for the image segmentation process in optimization algorithms. According to experiments and results, the improved Grasshopper algorithm is performs better compared to the optimization algorithms for Grasshopper, whale, firefly and bee colony. Keywords: Image segmentation, Improved Grasshopper Optimization Algorithm, Grasshopper Optimization Algorithm, Firefly optimization algorithm, Artificial Bee colony algorithm and Whale optimization algorithm. Abstract Images are the most important and widely used digital data used in computer systems. A digital image is made up of a set of objects or areas, so one of the efficient techniques for extracting features from images with respect to their constituent objects is the image segmentation technique, which delimits objects or areas. Highlights the image with high accuracy due to its texture and features. Using image segmentation, image pixels are placed next to each other in specific areas due to common features and generally similarity to each other. Multi-level image thresholding is one of the most popular and at the same time the simplest and most efficient methods of image segmentation. The most important issue in this method is the selection of the value of the relevant thresholds. In such a way that by determining the appropriate thresholds, the desired image can be more accurately zoned. Atsu method is one of the thresholding methods that has a good performance in determining two-level thresholds, but when increasing the number of thresholds, Atsu performance decreases in terms of time and segmentation accuracy. Therefore, it is combined with optimization algorithms to achieve better performance in terms of time and segmentation accuracy. In this research, an improved Grasshopper optimization algorithm is also proposed to increase the accuracy of finding answers and increasing the accuracy of segmentation, as well as to increase image quality. In this method, Atsu evaluation function is proposed for the image segmentation process in optimization algorithms. According to experiments and results, the improved Grasshopper algorithm is performs better compared to the optimization algorithms for Grasshopper, whale, firefly and bee colony. Keywords:

  Abstract:Today, using artificial intelligent and machine learning algorithms, we see improvements in intelligent tra  ortation industry, especially automatic vehicles that can analyze the information about them by using advanced sensors and machine vision techniques. The main challenge in design of this type of vehicles is identification of other vehicles around the vehicle’s path. The main objective of this thesis is to provide a method for identifying type of emergency vehicles based on deep learning. Considering the importance of passing emergency vehicles on roads and streets, non - drive vehicles should be able to identify this type of vehicles with high accuracy and have an appropriate response.In this thesis, to identify the type of emergency vehicles, methods based on deep learning are proposed which feature extraction and ltr">

   Skin cancer is one of the most common cancers in human societies and its prevalence is increasing dramatically. Melanoma is one of the most dangerous types of skin cancer, and the more the skin lesion grows, the lower the chance of cure. Early detection of cancer plays an important role in its treatment. Definitive treatment of melanoma cancer is possible with early detection. In this dissertation, a new method for diagnosing skin cancer was presented. In this method, two types of discrete and stationary wavelet transform were first applied to the images. A number of statistical features were then extracted from these converted images. Also, various global, local, etc. features were applied to the gray and color surface images. In the next step, to improve the results, the extraction features were combined to obtain the best combination of features that >Keywords: Melanoma Cancer, Discrete Wavelet Transform, Stationary Wavelet Transform, Least Squares Support Vector

طبقه­بندي كاني­ها بخش جدايي­ناپذيري از زمين­شناسي است. به‌صورت سنتي براي مطالعه­ كاني­هاي موجود در مقاطع نازك، مرز بين كاني­ها به‌صورت دستي جداشده، هر ناحيه برچسب­گذاري و درصد هر كاني محاسبه مي­شود. اين روش نيازمند دانش، تخصص و تجربه­ بالايي است. از سوي ديگر خطاي انساني ناشي از خستگي و بي­دقتي موجب كاهش دقت طبقه­بندي مي­شود. بنابراين به­كارگيري يك سامانه مبتني بر پردازش تصوير براي تشخيص خودكاركاني­هاي موجود در سنگ­ها امري ضروري است. ارائه چنين سامانه­اي مي­تواند باعث افزايش دقت، كاهش خطاهاي انساني، كاهش هزينه و كاهش زمان جهت تشخيص نوع كاني­ها مي­شود؛ بنابراين، هدف اين پژوهش، پيشنهاد يك سامانه تشخيص خودكار كاني است كه با استفاده از پردازش تصوير، كاني­هاي موجود در سنگ را شناسايي و طبقه­بندي كند. مرحله اول در انجام اين پژوهش ايجاد يك پايگاه داده از تصاوير مقاطع نازك سنگ است. اين مرحله يكي از چالش­برانگيزترين مراحل اين پژوهش بود، زيرا ايجاد يك پايگاه داده مناسب از تصاوير مقاطع نازك، فرآيندي سخت و وقت­گير است. از سوي ديگر، پايگاه داده مشترك و استانداردي در اين حوزه وجود ندارد و هر پژوهشي از پايگاه داده متفاوتي استفاده مي­كند. پس از ايجاد پايگاه داده و برچسب­گذاري تصاوير مقاطع نازك، چند روش­ قطعه­بندي بررسي و الگوريتم JSEG براي قطعه­بندي انتخاب شده است. پس از انجام قطعه­بندي، ويژگي­هاي مبتني بر رنگ و بافت از هر ناحيه استخراج شده­اند. ويژگي­هاي رنگي از هر دو فضاي رنگي RGB و HSI استخراج شده­اند. هم­چنين به دليل اينكه برخي كاني­هاي متفاوت داراي رنگ­هاي مشابه هستند، ويژگي­هاي بافت نيز از هر ناحيه استخراج شده­اند. ويژگي­هاي استخراج‌شده از هر ناحيه، براي طبقه­بندي به طبقه­بند فرستاده شده و طبقه­بند هر ناحيه را به‌عنوان يك كاني برچسب­گذاري كرده است. در اين پژوهش كارايي شش طبقه­بند Linear Discriminant، Su  ace Discriminant، Boosted Tree، Bagged Tree، Linear SVM و Weighted KNN بر اساس معيارهاي مختلف مورد ارزيابي قرار گرفته است. بر اساس نتايج تجربي به‌دست‌آمده، طبقه­بند Bagged Tree داراي بالاترين دقت به ميزان 5253/95 و همچنين كمترين ميزان خطاي MAE برابر با 0447/0 و خطاي RMSE برابر با 2115/0 مي­باشد. همچنين همه طبقه­بندها داراي دقت قابل قبول بالاي 93% هستند. اين نتايج نشان مي­دهد كه روش پيشنهادي داراي قابليت مناسبي جهت شناسايي خودكار كاني­هاست.  

   With advances in Internet technology and easy access to Internet networks, digital images that are available to the public can be modified, manipulated, and copied illegally without any qualitative loss. Without respecting the ownership of this data, the desired content can be reproduced and distributed on a large scale. Digital watermarking appears as a branch of data concealment science and protects the ownership and copyright of these images. In this dissertation, the idea of designing and implementing an invisible digital image watermarking method in the field of Lifting Wavelet Transform and Singular Value Decomposition in order to improve image Robustness and increase the reliability of image watermarking system using The Three Module Redundancy Technique is proposed. First, the Lifting Wavelet Transform is applied to the host image, and then its three high frequency su  ands divided into 8×8 non overlapping blocks. In each frequency subbands, the Lifting Wavelet Transform is converted to each block, and four frequency subbands ll, lh, hl and hh are obtained for each block. Then the lh frequency subband is selected from each block and the singular value Decomposition is applied to the selected frequency subband. In order to increase the security of the hidden image against alteration and manipulation, the hidden image is first encrypted using the improved Arnold Transform. Following the technique of the Three Module Redundancy, the Singular Values of the watermark image are inserted into the host image by adding the Singular Values of each frequency subbands. The three optimal scale factors used by the ant bee colony algorithm to determine the balance between Robustness and imperceptability are used to insert the watermark image. Evaluations show that the average hidden Robustness of different images against signal processing and geometric attacks is more than 96% and the average imperceptability of the hidden images is more than 50 dB. According to the assessments, the proposed method is more Robustness and imperceptability than the existing works, and the proposed system is a fault tolerance system that can work properly under different circumstances. Also, by using the improved Arnold conversion to encrypt the watermark image, attempt has been made to increase the image security.

different places (e.g. emergency rooms, bank branches, etc.) is long queues which cause them to waste their time. By designing a system that can provide people with information about the state of congestion in different places, if it used by people, it can help them to avoid wasting their time. This system can be very useful when citizens are asked not to be present in crowded places due to the spread of infectious diseases such as coronavirus. In this thesis, an IoT system based on edge computing architecture is designed to solve the issue of queue formation and population overcrowding. In this system, a Windows-based software receives images by connecting to wireless CCTV cameras and counts the number of people with the help of an image processing method, and then, sends the location status to the server. In this thesis, the background, studies conducted, and challenges of IoT-based people counting systems and image processing algorithms are discussed, and three image processing methods for counting people are proposed. Two methods are based on eliminating the background and counting people based on the background pixels, and the third method is a model based on the MRCNN networks, which is taught to count people’s heads. The places that this system is implemented to detect congestion in them are indoor places where there is rarely the problem of people’s shadows. The mean errors of the best method based on the MAE and RMSE scales in all the test frames obtained as 1.51 and 1.89, respectively.

The diagnosis of cancer is mainly performed by visual analysis of the pathologists, through examining the morphology of the tissue slices and the spatial arrangement of the cells under a microscope. If the microscopic image of a specimen is not stained, it will look colorless and textured. Therefore, chemical staining is required to create contrast and help identify specific tissue components. During tissue preparation due to differences in chemicals, scanners, cutting thicknesses, and laboratory protocols, similar tissues are usually varied significantly in appearance. This diversity in staining, in addition to interpretive disparity among pathologists more, is one of the main challenges in designing robust and flexible systems for automated analysis. Various strategies for normalizing stain have been proposed as a pre-processing step in automated pipeline systems. In this thesis, the stain normalization for Hematoxylin and Eosin (H&E) stained histopathology images has been performed based on the Pix2Pix framework derived from conditional generative adversarial networks (cGAN). The proposed approach is called "Stain-to-Stain Translation" (STST). This method learns not only the specific color distribution but also the preserves corresponding histopathological pattern. Also, unlike previous methods that depended on a reference image, this method uses the distribution of all images in the training set for learning. The STST method has achieved significant results, both quantitative and qualitative evaluation, against some of the best methods. Based on the obtained results, it can be shown that STST, besides the very high perceptual similarity between the ground truth and the restained image, outperformed other stain normalization methods examined on the processing time metric. It also in a clinical use-case, namely breast cancer tumor name="_ftnref1" title="">[1]. [1] https://github.com/pegahsalehi/Stain-to-Stain-Translation   

  كف هاي طاق ضربي متشكل از طاق هاي قوسي شكل بنايي با دهانه حدود يك متر هستند كه بارهاي ثقلي را به تكيه گاه هاي خود (تيرچه هاي فولادي) منتقل مي نمايند. تعداد قابل توجهي از كف هاي طاق ضربي در ساختمان هاي موجود   بدليل ناكافي بودن ممان اينرسي تيرچه هاي خود تحت بارهاي   پياده روي داراي مشكل ارتعاشات قائم بيش از حد هستند. با اين حال، مشكل ارتعاش تنها محدود به اين نوع كف نبوده و در ساختمان هاي جديد نيز   كه تمايل به استفاده از دهانه هاي بزرگ تر، كاهش پارتيشن ها و تيغه ها در ساختمان، و كاربرد مصالح سبك رايج است، كم و بيش مشكل ارتعاشات آزار دهنده كف هاي سازه اي مطرح مي باشد. در ساختمان هاي اداري جديد با توجه به   پيشرفت فناوري،   بسياري از كمدهاي مدارك و قفسه هاي بايگاني با كامپيوتري روميزي جايگزين شده اند. بنابراين، علاوه بر بارهاي مرده كف (بدليل سبك سازي)، بارهاي بهره برداري نيز نسبت به گذشته كاهش قابل توجهي يافته اند و   ميرايي موثر كف كاهش يافته است. اين عوامل سبب ايجاد ارتعاشات سازه اي بيش از حد در كف هاي مذكور شده است كه در درجه اول موجب سلب آسايش ساكنين شده و در بعضي موارد سبب اختلال در عملكرد تجهيزات حساس نصب شده در سازه مي گردد. يكي از راه حل هاي موجود جهت كاهش   ارتعاشات   افزايش سختي و فركانس سازه كف   با افزايش تعداد ستون ها و يا تغيير مقاطع تيرها و . . مي باشدكه اين امر نيازمند صرف زمان و هزينه زيادي است. به علاوه در اغلب موارد محدوديت هاي   معماري عمدتا مانع اين راهكار ميشوند. راه حل ديگر در كاهش ارتعاشات افزايش ميرايي موثر كف با استفاده از انواع ميراگرهاي   الحاقي مي باشد. در اين پايان نامه براي كاهش ارتعاشات   دو كف سازه اي مختلف (الف: طاق ضربي و ب: عرشه فولادي با ورق هاي موج دار) تحت بارگذاري هاي مختلف پياده روي   از ميراگرهاي جرمي تنظيم شونده ويسكوالاستيك استفاده بعمل آمده است. مدل سازي ها با استفاده از نرم افزار آباكوس انجام شده است.   مدل اجزا محدود ميراگرهاي مورد نظر براساس نتايج آزمايشگاهي ارتعاش آزاد پيشين كاليبره شده است. سپس، جانمايي مناسب ميراگر، با هدف دستيابي به بيش ترين كاهش در ارتعاشات تير، با مدلسازي ارتعاشات تيري با فركانس طبيعي 52/4 هرتز و كاربرد ميراگر   در 8 حالت مختلف مورد بررسي قرار گرفت. در ادامه با مدلسازي يك كف طاق ضربي و نيز يك كف عرشه فولادي داراي ارتعاشات بيش از حد مجاز، تاثير   ميراگرهاي با نسبت جرمي 1% ، 2%،   و 3% در كاهش ارتعاشات مورد بررسي قرار گرفت.   ارتعاشات كف هاي مذكورتحت اثر 6 بارگذاري پياده روي و 3 بارگذاري دويدن با فركانس هاي تحريك 5/1،   2، و 3 هرتز   بررسي شده است. نتايج اين تحقيق نشان مي دهد در حالتي كه ميراگر در وسط دهانه و موازي با تير قرار مي گيرد بيش ترين كاهش در شتاب حداكثر ايجاد شده در كف به دست مي آيد. چنانچه به هر دليل   امكان نصب پايه ميرا گر ها در وسط دهانه تيرچه كف ميسر نباشد، ميراگرها را بايد در طرفين مركز تيرچه بگونه اي نصب كرد كه جرم انتهايي آنها متمايل بسمت داخل دهانه تيرچه قرار گيرد تا كاهش حداكثري ايجاد شود. بر اساس مطالعات اين پايان نامه   ميراگرهاي جرمي تنظيم شونده مورد بررسي   قادر به كاهش ارتعاشات هر دو كف سازه اي تا   حد مجاز تحت اثر   بار هاي پياده روي   بوده و براي بارگذاري ناشي از دويدن تا 80% ارتعاشات را كاهش   مي دهند.  

Abstract Information diffusion prediction is the study of the path of dissemination of news, information, or topics in a structured data such as a graph. Research in this area is focused on two goals, tracing the information diffusion path and finding the members that determine future the next path. The major problem of traditional approaches in this area is the use of simple probabilistic methods rather than intelligent methods. Recent years have seen growing interest in the use of machine learning algorithms in this field. Recently, deep learning, which is a branch of machine learning, has been increasingly used in the field of information diffusion prediction. This paper presents a machine learning method based on the Graph Neural Network algorithm, which involves the selection of inactive vertices for activation based on the neighboring vertices that are active in a given scientific topic. Basically, in this method, information diffusion paths are predicted through the activation of inactive vertices by active vertices. The method is tested on three scientific bibliography datasets: DBLP, Pubmed, and Cora. The method attempts to answer the question that who will be the publisher of the next article in a specific field of science. The comparison of the proposed method with other methods shows 10% and 5% improved precision in DBLP and Pubmed datasets, respectively.   

  The main purpose of human data to collection is to understand the thinking of other human beings. This unconscious tendency has led researchers to analyze information in order to understand and analyze the minds of other human beings. Today, with the advancement of information platforms such as the Internet, social networks, etc., it is easy to gather the information you need. Today, social networks are one of the most important aspects of people's lives, and on the other hand, these networks have made huge profits by exploring the general information of users. The aim of this study is to investigate the text to find out the mood of people in typing texts. In this study, 14,000 tweets related to airlines were used to analyze emotions in three categories: positive, negative and neutral. The final proposal has three steps. In the first step, we perform the pre-processing operation on the database. In the second step, using the Imperialist Competitive Algorithm, we extract the main words from all the existing words. Keywords are the words that have the most impact on categorization. We then use the convolution neural network to extract more features. In the last step, we perform the classification operation using the multilayer perceptron neural network (MLP). At the end, using the final proposed design, we achieved precision, accuracy and recall of 0.990, 0.983 and 0.875, respectively. The results indicate that the final proposed design is desirable.

تهديدهاي امنيتي وب به­طور روزافزون در حال افزايش است. ماهيت شبكه اينترنت به صفحات وب بدخواه اين اجازه را مي­دهد تا خود را به‌عنوان "صفحات امن" نشان دهند و متعاقباً برخي از كاربراني كه آگاهي كافي ندارند در دام اين وب­سايت­ها گرفتار شوند. يكي از حملات رايج اين حوزه، حمله Cross-Site Scripting(XSS) است. اين حمله با تزريق اسكريپت­هاي مخرب به ورودي­هاي صفحات وب رخ مي­دهد، زماني كه كاربر صفحه آلوده مورد نظر را بازديد كند به وقوع مي­پيوندد. روش مرسوم براي شناسايي صفحات مخرب وب، استفاده از فهرست‌هاي سياه است. اين فهرست‌هاي سياه، توسط سازمان­هاي مورد اعتماد و داوطلب تهيه مي­شود و سپس توسط مرورگرهاي مدرن مانند كروم و فايرفاكس استفاده مي­شود. با توجه به اينكه، ماهيت صفحات وب به‌طور مداوم در حال تغيير است، اين روش در شناسايي تهديدهاي جديد ناكارآمد است رويكرد ديگر، استفاده از روش­هاي يادگيري ماشين است كه تصميم­گيري­هاي پيچيده‌تري نسبت به روش انساني مي­توانند اتخاذ كنند. روش­هاي يادگيري ماشين با تحليل ايستاي متن(بدون اجراي كد) اين كار را انجام مي­دهند اما هنوز هم عدم شناسايي صحيح در بسياري از برنامه­هاي جاري، منجر به فعال شدن كدهاي مخرب شده و آسيب وارد مي­كنند. در اين پژوهش هدف ما شناسايي وب­سايت­هاي مخرب با استفاده از تركيب تحليل ايستا و پوياي(با اجراي كد) است، كه به كمك اين دو رويكرد ابتدا، چالش­هاي رمزگشايي و مبهم­سازي را حل كرده و سپس ويژگي­هاي استخراج شده را تحليل مي­كنيم. نتايج اين تجزيه و تحليل نشان مي­دهد كه رويكرد پيشنهاد شده با الگوريتم طبقه­بندي درخت تصادفي، پيوندهاي صفحات وب را با دقت 97.11 درصد شناسايي مي­كند.   

   It is very time consuming to use deterministic methods for test generation as they use backtrack. The simulation-based test generation methods only analyze the circuit in forward path and this has made them popular. Random Test Generation Methods, which are among simulation-based methods, need a short time for test generation, but the number of test vectors produced in random methods is high. A suitable solution to reduce the number of these vectors is through using the Fault Coverage Index to evaluate the competency of test vectors and trimming test vectors that are inadequate. But calculating the Fault Coverage Index for each test vector requires a fault simulation that is a time consuming process. Also, the genetic algorithm can reach a very compact test set because of the optimized search it performs over a large space of test vectors. But this method, which is simulation based, again requires the time consuming simulation of fault as it uses the fault coverage index as a fitness function. The main purpose of this thesis is to reduce the test generation time in simulation-based methods by maintaining their quality for combinational circuits. The idea behind this thesis is to study the competency of test vectors using a new index based on Probabilistic  ystem that is fast and low-cost to calculate. To evaluate the accuracy of the proposed competency index, the concept of statistical correlation was used. The results showed that there is a correlation between the proposed competency index and the Fault Coverage Index for all circuits and the correlation was greater than 0.7 for 6 circuits out of 10 ISCAS85 circuits, which indicates high correlation.   The results of using the proposed competency index in simulation-based test generation methods showed that the basic method of trimming test vectors can be accelerated to 86% on average by maintaining the quality of test generation and the basic method of test generation based on genetic algorithm can be accelerated to 49.85% on average with an additional test vector.

 design interactive IoT smart toy based on Persian language

Abstract The advancement of technology and the Internet has led to an ever-increasing growth of databases, especially images, which has led to the search for the desired image and its recovery from the massive amount of databases. searching for images from the past has been an important research topic and several methods have been proposed, including methods for image retrieval based on text, the text-based retrieval method is a basic method and performs searches using the keywords defined for each image, given that the method of text search was a time consuming and costly method. attempts toward other methods and techniques, namely, image retrieval based on content, were made using descriptors of structural elements or low-level features of the image, ie, color, texture and shape, so that we can look at the search image. in this research, we have tried to describe the structural elements of SED and compare it with other descriptors and algorithms that are implemented in this implemented project and to achieve a higher degree of accuracy. by researching and investigating methods and descriptors of structural elements that utilize low-level features of color and texture, the proposed combination method is presented using structural elements and color difference histograms. on the other hand, considering that changing the size of images is an important issue and accessing the image with different sizes is considered an important issue, so the results of different methods of extracting features in 128× 128, 64× 64, 32× 32, 16× 16 and 8×

  The users and the common goals and objectives and put them in a batch, the proposed algorithm uses an distributed and interactive particle pooling algorithm. The distributed and interactive particle pool algorithm is a version of the PSO that can process each section of the database or each dimension of the target separately. Therefore, the proposed algorithm is well suited to distributed processing platforms such as Spark. The simulation results, while confirming the accuracy of the proposed method with the collaborative refinement, show that the proposed system for recommendation in the Kalandays is about 64 times faster than conventional processing platforms.

  Nowadays different approaches of signal processing are used in many applications due to its potential applicability to a wide range of problems, such as telecommunication and biomedical signals processing. Voice activity detection (VAD) is one of most important signal processing branches in audio signal processor and is used in many telecommunication systems such as Speech compression , speech recognition, upgrade of speech , noise estimation and noise removal. VADs are also used to detect input signals and >For instance in a mobile telecommunication system usually 60% of talk-time includes speech signal, so that the rest of the signal is not informative. To decrease channel capacity and power consumption in this case VADs can be used to submit only pure speech signal. There have been already many studies in this field, however the efficiency of proposed approaches are highly depends on background noise. So that their efficiency may decrease while noise power is quite higher compared to speech power. This current seminar aims to provide an efficient method which is based on the combination of typical detection techniques of the speech versus non-speech blocks, so that the result can be applied for both clean and high SNR environment

Today, the use of wireless technology has become very popular It has become an indelible part of everyday life and even the industry And an example of its applications can be cell phones, WiFi networks, Radio digital information exchange and more Cited.   To achieve these technologies, various engineering knowledge and achievements, especially in the field of electronic, have been used. Providing these features on a small chip in the advancement of the science of minimizing components VLSI , CMOS technology And the construction of RF components and circuits It's possible; RF design includes parts such as antennas, low noise amplifiers,mixers, oscillators, phase lock loops, frequency instruments and power amplifiers.An important part of the network after the antenna is the low noise amplifier, which has various parameters that interact with each other. Including noise, bandwidth, linearity, input matching and extra ; We have tried to get a more linear response from the circuit using the Linearity enhancement techniques and we have achieved 3.1-10.6 GHz bandwidth. The proposed circuit structure in the fourth chapter of this thesis is examined.key words:Low noise amplifier, bandwidth, linearity, gain, linearization technique

  mixed matrix membranes containing PEBAX, Glycerol Triacetate(GTA) and synthesized Alumina nanotubes(ANTs) was prepared by solution-casting method. The particles were synthesized by a hydrothermal method characterized using SEM, FTIR and XRD. Gas permeation test were applied for characterization and assessment of pure and resultant membranes. The synthesized nanotubes enhance the carbon dioxide pure gas permeability compared to pristine membrane. The effect of different loadings of ANTs on the permeability of CO2 and CH4 and ideal selectivity CO2/CH4 were investigated at the pressure of 5 Bar and temperature of 35°C. Then in the 4% loading of nanotubes, various amount of GTA(10-20-30-40 wt%) was added to the matrix which improved the gases permeability by disrupting of the chain packing and increasing the fraction free volume of the fabricated MMMs. Cross-sectional morphologies of membranes were characterize by   field emission scanning electron microscopy(FESEM). Fourier transform infrared (FT-IR) to identify variations of the chemical bonds were also applied. The result showed that for the fabricated membrane of pebax/4% wt ANTs/40 GTA, the permeability was enhanced almost 67% while the ideal selectivity decreased by 5.7%.

  Incorporation nanofillers and preparation of nanocomposites is an efficient and promising for improvement of the polymer matrixes different properties. In this study, halloysite nanotubes due to their instruction properties, good compatibility and it’s cheapness will incorporate in PEBA matrix for improving their separation performance for CO2/CH4. It’s expected that higher improvement can be achieved by modification the halloysite nanotubes for PEBA based nanocomposites. Optimally it is expected that the halloysite based nanocomposite membranes separation will be passed over the Robson’s upper bound as a measure of the membranes proper separation performance. Additionally, other membranes properties are expected to become better.

  This thesis examines the numerical and experimental performance of a metal foam heat sink in the free convection heat transfer. Metal foams are porcelain materials that used recently in the wide range application. This welcome is due to the appropriate thermo physical properties such as high volume ratio and high thermal conductivity. They are very lightweight because of their high porosity (0.9 and further).   The specimen used in this experiment is an aluminum foam (13 mm × 40 mm × 40 mm) of 92% porosity with 10 ppi. The experiments were carried out for heat sink inclination position of 0 °, 30 °, 60 °, 90 ° and 4, 8, 12, 16 watt power input. Numerical simulation was performed by finite element method and commercial software Comsol Multiphysics5.2. The heat transfer and fluid flow in the metal foam is expressed in terms of the macro volume theory based on the local thermal non-equilibrium condition (LTNE) for the energy equation. In this work, effect of foam geometric parameters, foam height, heat sink inclination angle and base temperature on the thermal performance of metal foam was investigated. The experiments results, show that the thermal performance of heat sink with increasing input heat flux decreases. For power input of 16 watts, the highest Nusselt number belong the inclination of 60 degrees (25.75).Comparison of thermal performance between horizontal and vertical heat sink indicates that the performance of horizontal heat sink is better than the vertical. The results of the numerical model show that the highest mean Nusselt number for all foam samples are in the horizontal position. It can be concluded that the average Nusselt number decreases with increasing porosity, and increases with decreasing in pore density. Influence of metal foam pore density of on the Nusselt number shows that in the samples with a 10 ppi and ppi 20 do not differ greatly, but the specimen with ppi 5 has a better performance. Comparison of Nusselt number in vertical(25.29) and horizontal(30.94) heat sink with 0.92% porosity and 5 ppi show that the mean Nusselt number in the horizontal position is %22.34 more than the vertical position. Comparison of Nusselt number of a metal foam heat sink ( 0.92% porosity and 10 ppi) with a flat plate in the horizontal position at the same base temperature (97.7 ?) indicated that the Aluminum metal foam increasing Nusselt number by %62.59. mean Nusselt number in the horizontal upward position is %29.52 more than the horizontal downward position

بيماري پاركينسون يكي از بيماري­هاي شايع عصبي است. اين بيماري با مشكلات حركتي براي بيماران همراه مي­باشد كه موجب عدم توانايي كاركردن و ديگر پيامدها مي­باشد. در اين پايان­نامه، سعي شده تصاوير مربوط به دست نوشته افرادي كه تست پاركينسون داده­اند به صورت اتوماتيك توسط روش­هاي پردازش تصاوير بررسي شوند و بيمارها و غير بيمار ها با متد­هاي پردازش ماشين و يادگيري ماشين تفكيك شوند. ويژگي­هاي الگوي باينري محلي و چندي­كردن فاز محلي براي اولين بار در مسئله­ي طبقه­بندي افراد سالم و بيمار پاركينسون بكار برده مي­شوند و پارامترهاي دقت شناسايي،   دقت،فراخواني وF-score   ارزيابي مي­شوند. روش پيشنهادي شامل سه قسمت است: پيش پردازش، استخراج ويژگي­ و كلاس بندي. در بخش پيش پردازش، نرمال سازي، قطعه­بندي مبتني بر عمليات ريخت­شناسي و فيلتر مات بر روي تصوير انجام مي­گردد. سپس، در بخش استخراج ويژگي براي تصوير، دست خط و خط چاپي از هم جدا شده و سپس با هم مقايسه مي­شوند تا ويژگي­هاي مربوط به آن به دست آيد. براي مشخص كردن نقاط متناظر روي دست خط و خط چاپي از اختلاف دو تصوير و همچنين ميانگين­گيري استفاده شده است. در ادامه، ويژگي­هاي بدست آمده كه مبتني بر اطلاعات آماري تصوير مي­باشد، بدست مي­آيد. در مرحله­ي بعد سه طبقه­بند مختلف ماشين بردار پشتيبان، نايو بيز و كا نزديك­ترين همسايه به منظور دسته بندي افراد سالم و بيمار پاركينسون بكار گرفته شده است. براي ارزيابي روش پيشنهادي و مقايسه با روش­هاي پيشين، از مجموعه داده Hand PD استفاده شده و از 90 درصد داده­ها براي آموزش و از 10 درصد براي تست استفاده كرده­ايم. نتايج به­دست آمده نشان مي­دهد كه بهترين الگوريتم در بين طبقه­بندها نايو بيز بوده است كه دقت اين روش براي طبقه­بندي افراد سالم و بيمار با بدست آوردن اطلاعات آماري   تصاوير، برابر با 32/85 است . همچنين در ادامه تاثير بكارگيري دو توصيفگر الگوي باينري محلي و الگوي چندي­ساز فاز محلي، بررسي شده است كه طبقه­بند نايوبيز بيشترين دقت را براي الگوي باينري محلي برابر با مقدار 77/87 و براي الگوي چندي­ساز فاز محلي برابر با 59/85   نتيجه داده است. در مجموع نتيجه­  hy  hy;ي بدست آمده از روش پيشنهاد شده نشان مي­دهد كه اين روش نسبت به روش­هاي اخير 9 درصد افزايش در دقت تشخيص داشته­است.   

يكي از مباحث مهم در جامعه امروزي كه دغدغه بسياري از كارشناسان و همچنين كاربران مي‌باشد بحث امنيت و تشخيص و تاييد هويت است. مردم خواستار اقدامات امنيتي بي­عيب، ساده و كاربرپسند هستند. بيومتريك، احراز هويت افراد براساس ويژگي­هاي منحصربفرد و متمايز كننده ، مقاوم و قابل­سنجش است كه بتواند جهت تعيين يا تأييد هويت افراد بكار رود. شناسايي از طريق بيومتريك، شناسايي يك فرد براساس صفات فيزيولوژي، رفتاري و شيميايي يك شخص است. تشخيص هويت از طريق بيومتريك مزاياي بسياري دارد و تاكنون روش­هاي مختلفي ارائه شده است. روش­هاي بكار رفته در هر دوره قوت و ضعف فناوري آن را به همراه دارد. در بين ويژگي­هاي بيومتريك مختلف استفاده از الگوي رگ دست افراد يكي از مناسب­ترين و قابل اطمينان­ترين خصيصه­هاي بيومتريكي مي­باشد كه ما در اين پايان­نامه به آن مي­پردازيم. سيستم­هاي تصديق هويت مبتني بر الگوي رگ دست شامل چندين مرحله مختلف از قبيل پيش­پردازش، استخراج ويژگي الگوي رگ­ها و تطابق الگو است. در سال­هاي اخير روش­هاي مختلفي براي هر كدام از اين مراحل ارائه شده است. در اين پايان نامه، تمركز ما بر روي استخراج ويژگي و بكارگيري توصيفگرهاي بافت تصوير و تركيب چند توصيفگر مي­باشد. به منظور استخراج ويژگي توصيفگرهاي الگوي باينري يكنواخت، الگوي باينري يكنواخت مستقل از چرخش و كوانتيزه ساز فاز محلي مستقل از چرخش به كار گرفته شده است. همچنين در روش پيشنهادي تركيب چند توصيفگر را نيز بررسي نموده ايم. در ادامه براي طبقه بندي تصاوير، سه طبقه بند متفاوت ماشين بردار پشتيبان، درخت تصميم و كا نزديك­ترين همسايه بكار گرفته شده است. براي ارزيابي دقيق روش پيشنهادي، از مجموعه داده PUT Hand Vein   كه خود شامل دو مجموعه داده از تصاوير رگ كف دست و رگ پشت دست است، استفاده شده است. پايگاه داده شامل 1200 تصوير رگ كف دست و همچنين 1200 تصوير رگ پشت دست است. همچنين پارامتر دقت طبقه بندي تصاوير و زمان محاسبات اندازه گيري شده است. نتايج بدست آمده از اجراي اين الگوريتم­ها و تركيبات مختلف آنها نشان مي­دهد كه بهترين الگوريتم تركيب الگوي باينري يكنواخت و كوانتيزه ساز فاز محلي است كه دقت اين روش در تصاوير رگ كف دست براي دست راست 99 درصد و براي دست چپ 33/ 99 درصد با طبقه­بند ماشين بردار پشتيبان بدست آمده است. در تصاوير رگ پشت دست براي دست راست مقدار دقت طبقه بندي 83/97 درصد و براي دست چپ 66/97 درصد با بكارگيري طبقه­بند كا نزديكترين همسايه بدست امده است. علاوه بر اين در مقايسه با روش هاي پيشين، نتايج بدست آمده از روش پيشنهادي بهبود دقت را نشان مي­دهد.  

  Investingation of the effect of thermodynamic and diffusion models on prediction accurately accurately of pervaporation separation performance

  AbstractThe most significant tool in combating crime is the fight against money laundering. Detecting “suspicious transactions of money laundering” in banks is the biggest challenge of combating money laundering. Lack of attention to the context of the bank account owner’s is resulting in low efficiency of anti-money laundering approaches. The aim of this study is to provide a method of detecting suspicious transactions based on data-mining techniques such as a statistical method for analyzing “contextual outlier transactions” by targeting money launderer’s transactions in integration stage. The method of this study was context analyzing, and its population includes 1.8 million simulated transactions belongs to 1008 people from 48 different contexts over a period of 6 years. Simulators probably distributions came from the Kolmogorov-Smirnov test on 50 person cross-sectional actual bank transactions sample. The transactions collected over field research. Due to the unavailability of sufficient numbers of actual bank transactions, simulated transactions used. By simulating, the ability to create scenarios that may not provide in the real world is possible. Testing the idea of the research resulted in 100% True Positive Rate and 1.14% False Positive Rate, compared to most methods, tangible progress achieved. The study findings showed attending to bank account owner’s context, promotes the quality of the methods used in detecting money laundry.  Keywords: Money-laundry, Context, Contextual Variable, Behavioral Variable, Working Set Window

Studies to date of the existence of a difference in peoples emotions. This vision for all researchers, particularly developers of computer games is valuable, Because by increasing the touchscreen and an increase in this type of game on our phones, the question arises, "Is touching behavior reflects the mood of the players?" If we can recognize the user’s emotions, according to the emotions of users, game design can control the amount and intensity of the game and to minimize the damaging effects of such games. . In this study, we characterized human touch in time on a touch screen use, So that we can distinguish between emotions and personality of users. In this study, using figures factor in diagnosing mental states were able to carefully 91/90 percent and 97/79 in the best position to do character recognition accuracy. In addition to this we got a result and it is other aspects of the recovery process does not recognize the characters in the parameters may in the algorithms of the parameters of feature in the evaluation of personality dimensions will be deleted when emotions are evaluated. But if consider arousal dimension moods and personality aspects to evaluating mood and personality dimensions also approached carefully 98/52 and will have a positive influence in results.

Augmented Reality (AR) applications rely on automatically matching a captured visual scene to an image in a database. The task of the thesis is to develop a technique which recognizes paintings displayed in an exhibition. Such a scheme would be useful as part of an electronic museum guide; the user would point his camera-phone at a painting of interest and would see/hear commentary based on the recognition result. Applications of this kind are usually referred to as "augmented reality" applications. Implemented on hand-held mobile devices, called "mobile augmented reality." We are interested in the image processing part of the problem.In this thesis, recognize image at the museum and a gallery is done. Photographed a database of Iraqi National museum and Free drawing exhibition in Ministry of culture and media in Baghdad. Recognize image evaluation parameters are time and accuracy. Features that are extracted from the images for the first time are Histogram in the different bin: histogram 256 bin, histogram 18 bin, and histogram 12 bin, Histogram of Oriented Gradients (HOG), Local Binary Patterns (LBP), Local configuration pattern (LCP). Also, these methods are compared with the three methods Scale Invariant Feature Transform (SIFT), Speed Up Robust Features (SURF), The combination of SIFT –SURF which has been used in past articles.The results showed that the best algorithms for image recognition are HOG-Histogram algorithm using SVM ltr">

  آثار هنري هر كشور از سرمايههاي ملي آن است كه حفظ اين سرمايهها امري مهم و ضروري است. تابلوهاينقاشي ترسيم شده توسط نقاشان هر كشور از مهمترين آثار هنري بشمار ميآيد. نقاشان معروف با ترسيم آثاريبرگرفته از ذهن خود تصوير جديدي از دنياي اطرافشان را به ديگران نشان ميدهند. جعل اين تابلوها از چالشهايياست كه همواره اصالت اثر هنري را زير سؤال ميبرد. ازاينرو تكنيكهاي آزمايشگاهي و خودكار براي كمك بهكارشناسان هنري مطرح ميشود. در دهه اخير با بكار بردن تكنيكهاي پردازش تصوير در اين زمينه موفقيتهايچشمگيري كسب شده است. استخراج ويژگيهايي كه منجر به طبقهبندي سبكهاي نقاشي ميشوند اولين قدمشناسايي آثار جعلي است.در اين پاياننامه طبقهبندي سبكهاي نقاشي بر روي دو پايگاه دادهي ايراني و غير ايراني انجام شده است.پايگاه دادهي ايراني شامل 320 نقاشي از نقاشان معروف ايراني بانامهاي بهزاد، كمالالملك، كاتوزيان، فرشچيان وسپهري است. پايگاه دادهي غير ايراني شامل 250 نقاشي از نقاشان معروف بانامهاي براك، مونه، ماتيس، پيكاسو وونگوگ است. ويژگيهاي الگوي باينري محلي، الگوي پيكربندي محلي، چندي كردن فاز محلي، تبديل ويژگيمقياس نابسته، تركيب ويژگيهاي هيستوگرام گراديان جهتدار و تبديل ويژگي مقياس نابسته و تركيب ويژگيالگوي پيكربندي محلي و هيستوگرام گراديان جهتدار براي اولين بار در مسئلهي طبقهبندي سبكهاي نقاشيبهكاربرده ميشوند و پارامترهاي دقت، زمان، انحراف معيار و مقاومت در برابر نويز ارزيابي ميشوند. همچنين روش-هاي مطرحشده را با سه روش هيستوگرام، هيستوگرام گراديان جهتدار و فيلتر گابور كه در ديگر مقالات بهكاربردهشدهاند مقايسه شدهاند.نتايج بهدستآمده نشان ميدهد كه بهترين الگوريتم تركيب هيستوگرام تصاوير رنگي با تبديل ويژگي مقياس98 و براي طبقهبندي / نابسته است كه دقت اين روش براي طبقهبندي سبكهاي نقاشي ايراني برابر با 1297 % است. در اين تحقيق براي شناسايي سبكهاي نقاشان ايراني افزايش / سبكهاي نقاشي غير ايراني برابر با 711 درصدي نسبت به روشهاي پيشين مشاهده / 3/5 درصدي و در شناسايي سبكهاي نقاشان غير ايراني افزايش 5ميشود.

 ارزيابي ويژگي هاي بافت تصاوير به منظور تعيين شكستگي استخوان