Iranian Society of Machine Vision and Image ProcessingJournal of Machine Vision and Image Processing2383-11978420211023Simultaneous determination of exchange rate and concentration of paramagnetic contrast agent in magnetic resonance imaging using electromagnetic pulse widthSimultaneous determination of exchange rate and concentration of paramagnetic contrast agent in magnetic resonance imaging using electromagnetic pulse width111129491FAMohamad-reza RezaeianBiomedical Engineering Department, Hamedan University of Technology, Hamedan, IranJournal Article20210422Imaging of the chemical exchange phenomenon by magnetic resonance method using saturation transfer process has made it possible to non-invasively diagnose some diseases through image processing knowledge. Imaging pulses consisting of a rectangular electromagnetic saturating pulse with two adjustable parameters of amplitude and width are commonly used. Saturation transfer depends on parameters such as rest times, chemical exchange rate, contrast agent concentration, and saturation pulse properties. Among these, the exchange rate and concentration of the contrast agent are much more effective due to their dependence on some clinical indicators such as pH, temperature and glucose consumption, so studies have been performed to measure them. The title asymmetric magnetization transfer is quantified in accordance with paramagnetic factors in an analytical method using parametric data derived from body tissue and data from experimental experiments. The accuracy of the objective function is evaluated by measuring the relative error resulting from the square of the difference between the squares of the denoised data taken from the objective function and the actual data (approximately 2%). Using the optimal pulse width in the form of a closed relation and the maximum objective function, the exchange rate and the concentration of the contrast agent are determined simultaneously.Imaging of the chemical exchange phenomenon by magnetic resonance method using saturation transfer process has made it possible to non-invasively diagnose some diseases through image processing knowledge. Imaging pulses consisting of a rectangular electromagnetic saturating pulse with two adjustable parameters of amplitude and width are commonly used. Saturation transfer depends on parameters such as rest times, chemical exchange rate, contrast agent concentration, and saturation pulse properties. Among these, the exchange rate and concentration of the contrast agent are much more effective due to their dependence on some clinical indicators such as pH, temperature and glucose consumption, so studies have been performed to measure them. The title asymmetric magnetization transfer is quantified in accordance with paramagnetic factors in an analytical method using parametric data derived from body tissue and data from experimental experiments. The accuracy of the objective function is evaluated by measuring the relative error resulting from the square of the difference between the squares of the denoised data taken from the objective function and the actual data (approximately 2%). Using the optimal pulse width in the form of a closed relation and the maximum objective function, the exchange rate and the concentration of the contrast agent are determined simultaneously.https://jmvip.sinaweb.net/article_129491_1617b7714d0808345ac3da43336a4970.pdfIranian Society of Machine Vision and Image ProcessingJournal of Machine Vision and Image Processing2383-11978420211023A survey of Datasets, Data Classification Algorithms, and Automatic detection of mental disorders Systems in Psychological Drawing TestA survey of Datasets, Data Classification Algorithms, and Automatic detection of mental disorders Systems in Psychological Drawing Test1336129851FAMaryam Fathi AhmadsaraeiPh.D. Student of Computer Engineering, Karaj Branch, Islamic Azad University, Karaj, IranAzam BastanfardDepartment of Computer Engineering, Karaj Branch, Islamic Azad University, Karaj, IranAmineh AminiDepartment of Computer Engineering, Karaj Branch, Islamic Azad University, Karaj, IranJournal Article20210501Handwriting, drawing, or patterns can be used to detect cognitive and personality disorders. Automating psychological tests can help to detect mental disorders early and prevent it from getting worse and with irreversible consequences. Automation of psychological drawing tests requires reviewing datasets of psychological drawing tests, reviewing image classification algorithms as an essential step, and finally reviewing the scoring of tests according to the standards. In this study, the recent researches in the field of datasets of drawing and handwritten tests, various methods of classification of drawing tests, scoring techniques to the tests and challenges ahead have been thoroughly investigated. So far, no such comprehensive research has been conducted on datasets, classification algorithms, and automatic detection in psychological drawing tests. Also, a comprehensive comparison of how datasets are collected, drawing test classification methods, test comparison techniques with the standards, advantages, and disadvantages of each method is presented. Challenges in the process of automatic diagnosis in psychological drawing tests are also discussed. This study aims to identify fast, accurate methods with low processing load and high reliability. In this study, by comparing the presented methods, it is concluded that in classifying images and detecting mental disorders, neural network algorithms have higher accuracy than machine learning algorithms, but are slower.Handwriting, drawing, or patterns can be used to detect cognitive and personality disorders. Automating psychological tests can help to detect mental disorders early and prevent it from getting worse and with irreversible consequences. Automation of psychological drawing tests requires reviewing datasets of psychological drawing tests, reviewing image classification algorithms as an essential step, and finally reviewing the scoring of tests according to the standards. In this study, the recent researches in the field of datasets of drawing and handwritten tests, various methods of classification of drawing tests, scoring techniques to the tests and challenges ahead have been thoroughly investigated. So far, no such comprehensive research has been conducted on datasets, classification algorithms, and automatic detection in psychological drawing tests. Also, a comprehensive comparison of how datasets are collected, drawing test classification methods, test comparison techniques with the standards, advantages, and disadvantages of each method is presented. Challenges in the process of automatic diagnosis in psychological drawing tests are also discussed. This study aims to identify fast, accurate methods with low processing load and high reliability. In this study, by comparing the presented methods, it is concluded that in classifying images and detecting mental disorders, neural network algorithms have higher accuracy than machine learning algorithms, but are slower.https://jmvip.sinaweb.net/article_129851_7edbb84aa736f11be7714f905472246e.pdfIranian Society of Machine Vision and Image ProcessingJournal of Machine Vision and Image Processing2383-11978420211023Ranking and selecting the appropriate filter for Gaussian noise removal with TOPSIS methodRanking and selecting the appropriate filter for Gaussian noise removal with TOPSIS method3750130323FAMehrdad NabahatPh.D Student of Mathematics, Tabriz Branch, Islamic Azad University, Tabriz, Iran.0000-0002-4930-4086Farzin Modarres KhiyabaniDepartment of Mathematics, Tabriz Branch, Islamic Azad University Tabriz, IranNima Jafari NavimipourDepartment of Computer engineering, Tabriz Branch, Islamic Azad University Tabriz, Iran0000-0002-5514-5536Journal Article20210510Image noise removal is very important in most image processing analysis. Among various noises that can damage the image, Gaussian noise is one of the most notable. There are several ways to reduce image noises, including transformation-based methods, filter-based methods, and non-local methods.In most of the methods, only one criterion (i.e. signal-to-noise ratio) is used to reduce the image noise, and even some methods do not retain image quality and structural details of the image such as lines and edges.In this paper, the TOPSIS method was used to rank different filters of Gaussian noise filtering based on various criteria that influence the evaluation of image quality.In fact, regarding the standard deviation degree of the Gaussian noise, it was intended to know what kind of noise removal filter which contains the considered criteria should be used. The computational results of the proposed method were evaluated on different images with various standard deviation of Gaussian noise, and consequently, the filter has been specified corresponding to the degree of Gaussian noise for the application.Image noise removal is very important in most image processing analysis. Among various noises that can damage the image, Gaussian noise is one of the most notable. There are several ways to reduce image noises, including transformation-based methods, filter-based methods, and non-local methods.In most of the methods, only one criterion (i.e. signal-to-noise ratio) is used to reduce the image noise, and even some methods do not retain image quality and structural details of the image such as lines and edges.In this paper, the TOPSIS method was used to rank different filters of Gaussian noise filtering based on various criteria that influence the evaluation of image quality.In fact, regarding the standard deviation degree of the Gaussian noise, it was intended to know what kind of noise removal filter which contains the considered criteria should be used. The computational results of the proposed method were evaluated on different images with various standard deviation of Gaussian noise, and consequently, the filter has been specified corresponding to the degree of Gaussian noise for the application.https://jmvip.sinaweb.net/article_130323_99ea819b44a29a888f34b7ee6a31f2f3.pdfIranian Society of Machine Vision and Image ProcessingJournal of Machine Vision and Image Processing2383-11978420211023A review on deep learning approaches for optical character recognition with emphasis on Persian, Arabic and Urdu scriptsA review on deep learning approaches for optical character recognition with emphasis on Persian, Arabic and Urdu scripts5185132419FAShima KashefVaje Research Group, Kerman, IranHossein Nezamabadi-pourDepartment of Electrical Engineering, Shahid Bahonar University of Kerman, Kerman, IranElham ShabaniniaDepartment of Computer Engineering, Sirjan University of Technology, Sirjan, IranJournal Article20210624In recent years, the success of deep convolutional neural networks in object recognition has attracted the attention of many different areas of machine learning, including the field of optical character recognition, to this category. One of the major challenges in this area is to extract distinctive and informative features. Most of the methods proposed in the optical recognition of letters in recent years are based on hand-crafted features that have limited generalizability. Today, with the help of convolutional networks, feature extraction can be left to the machine automatically and with high efficiency. Also, structures based on the combination of convolutional and recursive networks have been proposed, which can perform recognition without the need for letter separation. This approach has received a great deal of attention from machine vision researchers in recent years; since, with the help of these networks, recognition can be done independently of the language and only according to the training set. The purpose of this article is to review the work done with this new approach in the field of optical character recognition. To this end, after stating the problem and a brief overview of the previous methods, methods based on deep learning algorithms and their characteristics are evaluated in more detail. Since the emphasis of this article is on research on optical recognition of letters in continuous scripts, such as Persian, Arabic and Urdu, the work done in these areas is also reviewed in a separate section. Also, while introducing famous datasets for different applications and reviewing the evaluation criteria of optical character recognition methods, the most important software and open source packages that are used for optical character recognition will be introduced.In recent years, the success of deep convolutional neural networks in object recognition has attracted the attention of many different areas of machine learning, including the field of optical character recognition, to this category. One of the major challenges in this area is to extract distinctive and informative features. Most of the methods proposed in the optical recognition of letters in recent years are based on hand-crafted features that have limited generalizability. Today, with the help of convolutional networks, feature extraction can be left to the machine automatically and with high efficiency. Also, structures based on the combination of convolutional and recursive networks have been proposed, which can perform recognition without the need for letter separation. This approach has received a great deal of attention from machine vision researchers in recent years; since, with the help of these networks, recognition can be done independently of the language and only according to the training set. The purpose of this article is to review the work done with this new approach in the field of optical character recognition. To this end, after stating the problem and a brief overview of the previous methods, methods based on deep learning algorithms and their characteristics are evaluated in more detail. Since the emphasis of this article is on research on optical recognition of letters in continuous scripts, such as Persian, Arabic and Urdu, the work done in these areas is also reviewed in a separate section. Also, while introducing famous datasets for different applications and reviewing the evaluation criteria of optical character recognition methods, the most important software and open source packages that are used for optical character recognition will be introduced.https://jmvip.sinaweb.net/article_132419_dc667aaa14b6ea1c987873d5e9d5adbe.pdfIranian Society of Machine Vision and Image ProcessingJournal of Machine Vision and Image Processing2383-11978420211023Subject adaptation in spontaneous facial behavior recognition using style transfer functionsSubject adaptation in spontaneous facial behavior recognition using style transfer functions8798133366FAAmin MohammadianBio Signal Processing Group, Research Center (RCDAT)Hassan AghaeiniaDepartment of Electrical Engineering Communications, Amirkabir University of technology, TehranFarzad TowhidkhahBiomedical engineering, Amirkabir university of technologyJournal Article20210711The appearance of facial Action Units (AUs) and painful expression may significantly vary for different people. Thus the probability distribution of both test and training data is not the same for person-independent facial behavior recognition. Some researchers have proposed methods to bring the performance of a person-independent system closer to a person-dependent one. Subject style is the cause of inter-personal variations. With this in mind, we propose methods to increase the generalization ability of facial AUs and pain detection through style transfer functions. We conducted extensive experiments on spontaneous UNBC-McMaster database to compare supervised methods. The results show that our approach can effectively perform the task of pain and AUs detection. So that the best average recognition rate of action units was 96.84 (with AUC criterion) and the same method in terms of low adaptation data and appropriate adaptation data had pain recognition rates of 87.30 and 93.26 (with AUC criterion), respectively.The appearance of facial Action Units (AUs) and painful expression may significantly vary for different people. Thus the probability distribution of both test and training data is not the same for person-independent facial behavior recognition. Some researchers have proposed methods to bring the performance of a person-independent system closer to a person-dependent one. Subject style is the cause of inter-personal variations. With this in mind, we propose methods to increase the generalization ability of facial AUs and pain detection through style transfer functions. We conducted extensive experiments on spontaneous UNBC-McMaster database to compare supervised methods. The results show that our approach can effectively perform the task of pain and AUs detection. So that the best average recognition rate of action units was 96.84 (with AUC criterion) and the same method in terms of low adaptation data and appropriate adaptation data had pain recognition rates of 87.30 and 93.26 (with AUC criterion), respectively.https://jmvip.sinaweb.net/article_133366_33bb7e527d270e3660623f37d6210b8f.pdfIranian Society of Machine Vision and Image ProcessingJournal of Machine Vision and Image Processing2383-11978420211023Identifying the reliability of the skeletal joints and estimating the position ofmissed joints in the Kinect sensor data to improve the diagnosis of musculoskeletal disordersIdentifying the reliability of the skeletal joints and estimating the position ofmissed joints in the Kinect sensor data to improve the diagnosis of musculoskeletal disorders99110134017FAAtiye MirmoiniRIV Lab., Department of Computer Engineering, Bu Ali Sina University, Hamedan, IranHassan KhotanlouRIV Lab., Department of Computer Engineering, Bu Ali Sina University, Hamedan, IranVahid PouraminDepartment of Computer Engineering, Sayyed Jamaleddin Asadabadi University, Asadabad, IranElham AlighardashDepartment of Computer Engineering, Sayyed Jamaleddin Asadabadi University, Asadabad, IranJournal Article20210723Applying Kinect sensorsare increasing in recent years due to low price and wide applications. This device can estimate the posture of the human body by utilization of skeletal data and without using any marker. Obstruction of the human body by other objects and rapid movement in front of the Kinect are the main problems in estimating the position of the joints. Two steps are considered in this study to solve the existing challenge.In the first step, a solution based on measurement models has been proposed to determine the reliability of joint position that extracted from Kinect, which is considered as an effective feature associated with the joint position in the Max-Margin classifier. Then, based on the reliability of each joint, a decision is made and the missing joints are identified. Finally, the joints are accredited using human body segmentation algorithms based on a deep learning network. The results show that selection of appropriate features in the first step to verify consecutive frames compared to existing approaches,has a significant improvement in the accuracy of the classification. The second phase also has a supreme impact on the accuracy of the methods that use the Kinect sensor skeletal data as input features by applying validation.Applying Kinect sensorsare increasing in recent years due to low price and wide applications. This device can estimate the posture of the human body by utilization of skeletal data and without using any marker. Obstruction of the human body by other objects and rapid movement in front of the Kinect are the main problems in estimating the position of the joints. Two steps are considered in this study to solve the existing challenge.In the first step, a solution based on measurement models has been proposed to determine the reliability of joint position that extracted from Kinect, which is considered as an effective feature associated with the joint position in the Max-Margin classifier. Then, based on the reliability of each joint, a decision is made and the missing joints are identified. Finally, the joints are accredited using human body segmentation algorithms based on a deep learning network. The results show that selection of appropriate features in the first step to verify consecutive frames compared to existing approaches,has a significant improvement in the accuracy of the classification. The second phase also has a supreme impact on the accuracy of the methods that use the Kinect sensor skeletal data as input features by applying validation.https://jmvip.sinaweb.net/article_134017_7bd0676ee0835e6bd9bf8c58f8bb01f8.pdfIranian Society of Machine Vision and Image ProcessingJournal of Machine Vision and Image Processing2383-11978420211023Development of real-time machine vision system for water balance monitoring of Lilium flowerDevelopment of real-time machine vision system for water balance monitoring of Lilium flower111127140325FAHadis BiabiMSc student of Mechanics of Biosystems Engineering, Agricultural Engineering and Rural Development, Khuzestan Agricultural Sciences and Natural Resources UniversitySaman Abdanan MehdizadehAgricultural Engineering and Rural Development, Agricultural Sciences and Natural Resources University of KhuzestanHadi OrakMSc student of Mechanics of Biosystems Engineering, Agricultural Engineering and Rural Development, Khuzestan Agricultural Sciences and Natural Resources University0000-0002-5873-802XMohammadreza Salehi SalmiHorticultural Science Department, of Agricultural, Khuzestan Agricultural Sciences and Natural Resources UniversityJournal Article20211118Water plays an important role in the growth and health of plants. The supply of this vital material affects not only agricultural yields, but also the quality and control of their physiological processes. Therefore, water scarcity has a negative effect on the growth of plants. In this research, in order to detect the plant's water requirement, a series of images of lilium plant under drought stress conditions were investigated to extract the color and morphological features and based on them, an intelligent system was designed to recognize the plant's water status. After analyzing the parameters extracted from the imagesusing statistical analysis at 5% probability level and the sequential forward feature ranking method, the most suitable features were selected to predict the moisture content of the plant; Furthermore, the classification was carried out using the support vector machine (SVM) with different kernels. Finally, it was shown that the classification accuracy for the linear kernels, sigmoeid, and RBF was with 6, 9, and 9 features, respectively, were 81.19, 81.04 and 83.12%, respectively. The results showed that the proposed system had the ability to determine the levels of stress and control the amount of water neededWater plays an important role in the growth and health of plants. The supply of this vital material affects not only agricultural yields, but also the quality and control of their physiological processes. Therefore, water scarcity has a negative effect on the growth of plants. In this research, in order to detect the plant's water requirement, a series of images of lilium plant under drought stress conditions were investigated to extract the color and morphological features and based on them, an intelligent system was designed to recognize the plant's water status. After analyzing the parameters extracted from the imagesusing statistical analysis at 5% probability level and the sequential forward feature ranking method, the most suitable features were selected to predict the moisture content of the plant; Furthermore, the classification was carried out using the support vector machine (SVM) with different kernels. Finally, it was shown that the classification accuracy for the linear kernels, sigmoeid, and RBF was with 6, 9, and 9 features, respectively, were 81.19, 81.04 and 83.12%, respectively. The results showed that the proposed system had the ability to determine the levels of stress and control the amount of water neededhttps://jmvip.sinaweb.net/article_140325_fce81a747f9b6d5c550b3c29c3ae58a0.pdf