Pub Date : 2022-02-23DOI: 10.1109/MVIP53647.2022.9738775
Maziyar Grami
In recent years, human-computer interaction and machine vision have become two of the favorite research areas in computer science. This paper is research on hand motion and hand gesture recognition using image processing techniques to control some system commands. Different hand motion and gesture recognition methods have been considered by researchers for use in computer systems, video game consoles, and mobile devices. In such cases, hand motion or hand gesture type is detected by tracking the hands' image or by matching the image with gestures storing in the database, after the first position of the hand is identified. In this paper, we provide an efficient way for recognizing and tracking the sequences of image frames. The main objective of this paper is to provide an efficient method for hand recognition in crowded environments, without any restrictions. The frames can be received from a video file. In the first frame, the location of the hand is recognized using color analysis, then it would be compared to the next frames. Detected movements are applied for controlling commands in an operating system. So far, an effective way for this kind of problem has not been suggested. In this study, we have tried to propose an efficient algorithm that is less dependent on background and lighting conditions. We tested the proposed method on several captured videos using the Image Processing Toolbox of MATLAB. These videos are captured in very crowded environments. The presented results showed that in different lighting conditions and various noises, the proposed method works almost without mistakes.
{"title":"A New Algorithm for Hand Gesture Recognition in Color Videos for Operating System Commands","authors":"Maziyar Grami","doi":"10.1109/MVIP53647.2022.9738775","DOIUrl":"https://doi.org/10.1109/MVIP53647.2022.9738775","url":null,"abstract":"In recent years, human-computer interaction and machine vision have become two of the favorite research areas in computer science. This paper is research on hand motion and hand gesture recognition using image processing techniques to control some system commands. Different hand motion and gesture recognition methods have been considered by researchers for use in computer systems, video game consoles, and mobile devices. In such cases, hand motion or hand gesture type is detected by tracking the hands' image or by matching the image with gestures storing in the database, after the first position of the hand is identified. In this paper, we provide an efficient way for recognizing and tracking the sequences of image frames. The main objective of this paper is to provide an efficient method for hand recognition in crowded environments, without any restrictions. The frames can be received from a video file. In the first frame, the location of the hand is recognized using color analysis, then it would be compared to the next frames. Detected movements are applied for controlling commands in an operating system. So far, an effective way for this kind of problem has not been suggested. In this study, we have tried to propose an efficient algorithm that is less dependent on background and lighting conditions. We tested the proposed method on several captured videos using the Image Processing Toolbox of MATLAB. These videos are captured in very crowded environments. The presented results showed that in different lighting conditions and various noises, the proposed method works almost without mistakes.","PeriodicalId":184716,"journal":{"name":"2022 International Conference on Machine Vision and Image Processing (MVIP)","volume":"134 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124150765","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-02-23DOI: 10.1109/MVIP53647.2022.9738554
Seyed Amir Mousavi, M. Ahmadzadeh, Ehsan Yazdian
Magnetic Resonance Imaging as non-invasive imaging can produce detailed anatomical images. MRI is a time- consuming imaging technique. Several imaging techniques, like parallel imaging, have been suggested to enhance imaging speed. Compressive Sensing MRI utilizes the sparsity of MR images to reconstruct MR images with under-sampled k-space data. It has already been shown that convolutional neural networks work better than sparsity-based approaches in image quality and reconstruction speed. In this paper, a novel method based on very deep CNN for the reconstruction of MR images is proposed using Generative Adversarial Networks. Generative and discriminative networks are designed with improved ResNet architecture. Using improved architecture has led to deepening generative and discriminative networks, reducing aliasing artifacts, more accurate reconstruction of edges, and better reconstruction of tissues. Compared to DLMRI and DAGAN methods, we demonstrate the proposed method outperforms the conventional methods and deep learning-based approaches. Assessment is made on several datasets such as the brain, heart, and prostate. Reconstruction of brain data with a Cartesian mask of 30% in the proposed method has improved the SSIM criteria up to 0.99. Also, image reconstruction time is approximately 20 ms on GPU, which is suitable for real-time applications.
{"title":"Compressed Sensing MRI Reconstruction Using Improved U-net based on Deep Generative Adversarial Networks","authors":"Seyed Amir Mousavi, M. Ahmadzadeh, Ehsan Yazdian","doi":"10.1109/MVIP53647.2022.9738554","DOIUrl":"https://doi.org/10.1109/MVIP53647.2022.9738554","url":null,"abstract":"Magnetic Resonance Imaging as non-invasive imaging can produce detailed anatomical images. MRI is a time- consuming imaging technique. Several imaging techniques, like parallel imaging, have been suggested to enhance imaging speed. Compressive Sensing MRI utilizes the sparsity of MR images to reconstruct MR images with under-sampled k-space data. It has already been shown that convolutional neural networks work better than sparsity-based approaches in image quality and reconstruction speed. In this paper, a novel method based on very deep CNN for the reconstruction of MR images is proposed using Generative Adversarial Networks. Generative and discriminative networks are designed with improved ResNet architecture. Using improved architecture has led to deepening generative and discriminative networks, reducing aliasing artifacts, more accurate reconstruction of edges, and better reconstruction of tissues. Compared to DLMRI and DAGAN methods, we demonstrate the proposed method outperforms the conventional methods and deep learning-based approaches. Assessment is made on several datasets such as the brain, heart, and prostate. Reconstruction of brain data with a Cartesian mask of 30% in the proposed method has improved the SSIM criteria up to 0.99. Also, image reconstruction time is approximately 20 ms on GPU, which is suitable for real-time applications.","PeriodicalId":184716,"journal":{"name":"2022 International Conference on Machine Vision and Image Processing (MVIP)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132035585","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-02-23DOI: 10.1109/MVIP53647.2022.9738737
Zahra Hossein-Nejad, M. Nasri
In this article, a new approach is suggested in remote-sensing images registration. In the suggested approach, first, the features extraction process is done based on proposed redundant keypoint elimination method synthetic aperture radar-SIFT (RKEM-SARSIFT). Second, creating descriptors is based on the Multi-Support Region Order-Based Gradient Histogram (MROGH) algorithm. Finally, matching process is done based on nearest neighbor distance ratio (NNDR) and transformation model is done based affine transform. The simulation results on several remote sensing image datasets affirm the suggested approach advantage in comparison with some other basic registration methods in terms of precision matching, SITMMR and SITMMC.
{"title":"Image Registration Based on Redundant Keypoint Elimination SARSIFT Algorithm and MROGH Descriptor","authors":"Zahra Hossein-Nejad, M. Nasri","doi":"10.1109/MVIP53647.2022.9738737","DOIUrl":"https://doi.org/10.1109/MVIP53647.2022.9738737","url":null,"abstract":"In this article, a new approach is suggested in remote-sensing images registration. In the suggested approach, first, the features extraction process is done based on proposed redundant keypoint elimination method synthetic aperture radar-SIFT (RKEM-SARSIFT). Second, creating descriptors is based on the Multi-Support Region Order-Based Gradient Histogram (MROGH) algorithm. Finally, matching process is done based on nearest neighbor distance ratio (NNDR) and transformation model is done based affine transform. The simulation results on several remote sensing image datasets affirm the suggested approach advantage in comparison with some other basic registration methods in terms of precision matching, SITMMR and SITMMC.","PeriodicalId":184716,"journal":{"name":"2022 International Conference on Machine Vision and Image Processing (MVIP)","volume":"88 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131776224","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-02-23DOI: 10.1109/MVIP53647.2022.9738765
Abbas Barhoun, A. M. Khiavi, Alireza Sokhandan Sorkhabi, H. S. Aghdasi, Behzad Kargari
Froth flotation is one of the most important and widespread methods of separation of minerals and waste materials and at the same time one of the most accurate methods of refining low-grade metal minerals. This paper presents a method for visual feature extraction of froth bubbles including the size, color, shape, and mobility based on machine vision and image processing techniques. The proposed method is capable of identifying bubbles properties as well as estimating their velocity and direction of movement. The performance of the proposed method is evaluated using real videos captured from the copper floatation process. The method description, as well as simulation results, are presented.
{"title":"A Machine Vision Based Method for Extracting Visual Features of Froth in Copper Floatation Process","authors":"Abbas Barhoun, A. M. Khiavi, Alireza Sokhandan Sorkhabi, H. S. Aghdasi, Behzad Kargari","doi":"10.1109/MVIP53647.2022.9738765","DOIUrl":"https://doi.org/10.1109/MVIP53647.2022.9738765","url":null,"abstract":"Froth flotation is one of the most important and widespread methods of separation of minerals and waste materials and at the same time one of the most accurate methods of refining low-grade metal minerals. This paper presents a method for visual feature extraction of froth bubbles including the size, color, shape, and mobility based on machine vision and image processing techniques. The proposed method is capable of identifying bubbles properties as well as estimating their velocity and direction of movement. The performance of the proposed method is evaluated using real videos captured from the copper floatation process. The method description, as well as simulation results, are presented.","PeriodicalId":184716,"journal":{"name":"2022 International Conference on Machine Vision and Image Processing (MVIP)","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129833643","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-02-23DOI: 10.1109/MVIP53647.2022.9738778
Zohoor Hayali, G. Akbarizadeh
In sugar factories, crystal particle analysis plays an important role in the quality of sugar production. Analyzes include measuring the crystals' dimensions, estimating the area and distribution of the crystals in terms of dimensions, which are of great help in setting up sugar kilns. To be able to analyze sugar particles, we must first be able to segment crystals correctly. Therefore, segmentation is the first and most important stage of the analysis. This paper introduces a method based on Transfer Learning (TL) in deep neural networks for the Semantic Segmentation of sugar crystals. In this method, by modifying a pre-trained Convolutional Neural Network (CNN) called DeepLab, the semantic Segmentation of sugar crystals is performed and the results clearly show that this method can label the crystals with high accuracy and remove extra parts.
{"title":"Transfer Learning on Semantic Segmentation for Sugar Crystal Analysis","authors":"Zohoor Hayali, G. Akbarizadeh","doi":"10.1109/MVIP53647.2022.9738778","DOIUrl":"https://doi.org/10.1109/MVIP53647.2022.9738778","url":null,"abstract":"In sugar factories, crystal particle analysis plays an important role in the quality of sugar production. Analyzes include measuring the crystals' dimensions, estimating the area and distribution of the crystals in terms of dimensions, which are of great help in setting up sugar kilns. To be able to analyze sugar particles, we must first be able to segment crystals correctly. Therefore, segmentation is the first and most important stage of the analysis. This paper introduces a method based on Transfer Learning (TL) in deep neural networks for the Semantic Segmentation of sugar crystals. In this method, by modifying a pre-trained Convolutional Neural Network (CNN) called DeepLab, the semantic Segmentation of sugar crystals is performed and the results clearly show that this method can label the crystals with high accuracy and remove extra parts.","PeriodicalId":184716,"journal":{"name":"2022 International Conference on Machine Vision and Image Processing (MVIP)","volume":"26 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114034757","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-02-23DOI: 10.1109/MVIP53647.2022.9738750
Ali Dadgar, Y. Baleghi, M. Ezoji
In this paper a method is presented to estimate the position of feet/head of objects in various camera views. In this method, first, all objects in the scene are detected using the background subtraction. Then, human and non-human objects are separated via the support vector machine (SVM) that is trained based on local binary patterns (LBP) features. The basic idea of the next step of this work is that the feet/head of an object are the group of pixels that are projected to small region on ground/top plane by corresponding homography matrix. This idea is expressed via an optimization problem which avoids partitioning out small group of pixels. Experimental results show that the proposed methods can improve the accuracy of the object tracking.
{"title":"Improvement of Human Tracking Based on an Accurate Estimation of Feet or Head Position","authors":"Ali Dadgar, Y. Baleghi, M. Ezoji","doi":"10.1109/MVIP53647.2022.9738750","DOIUrl":"https://doi.org/10.1109/MVIP53647.2022.9738750","url":null,"abstract":"In this paper a method is presented to estimate the position of feet/head of objects in various camera views. In this method, first, all objects in the scene are detected using the background subtraction. Then, human and non-human objects are separated via the support vector machine (SVM) that is trained based on local binary patterns (LBP) features. The basic idea of the next step of this work is that the feet/head of an object are the group of pixels that are projected to small region on ground/top plane by corresponding homography matrix. This idea is expressed via an optimization problem which avoids partitioning out small group of pixels. Experimental results show that the proposed methods can improve the accuracy of the object tracking.","PeriodicalId":184716,"journal":{"name":"2022 International Conference on Machine Vision and Image Processing (MVIP)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133155244","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-02-23DOI: 10.1109/MVIP53647.2022.9738736
M. Noei, Mohammadreza Parvizimosaed, Aliakbar Saleh Bigdeli, Mohammadmostafa Yalpanian
Pneumonia is a life-threatening and prevalent disease and needs to be diagnosed within a short time because of the lungs' fluid flow. Late detection of the disease may result in the patient’s death. Thus, advanced diagnosis is a critical factor besides the disease progress. In addition to advanced diagnosis, the privacy of datasets is important for organizations. Due to the great value of datasets, hospitals do not want to share their datasets, but they want to share their trained network weights. Therefore, in this paper, we combine deep learning and blockchain to implement the blockchain as distributed storage. Using permission blockchain, weights are broadcasted among other hospitals securely. Because of the security, the dataset is shared with five hospitals equally. Each hospital trains its network model and sends its weights to the blockchain. The goal is to broadcast the aggregated weights among hospitals securely and have good enough results because the whole dataset is not implemented to train a network. The dataset contains 5856 images, and hospitals implement a residual neural network with 28 layers. The results show that hospitals can increase the accuracy of their model using shared weights compared to a model without using shared weights.
{"title":"A Secure Hybrid Permissioned Blockchain and Deep Learning Platform for CT Image Classification","authors":"M. Noei, Mohammadreza Parvizimosaed, Aliakbar Saleh Bigdeli, Mohammadmostafa Yalpanian","doi":"10.1109/MVIP53647.2022.9738736","DOIUrl":"https://doi.org/10.1109/MVIP53647.2022.9738736","url":null,"abstract":"Pneumonia is a life-threatening and prevalent disease and needs to be diagnosed within a short time because of the lungs' fluid flow. Late detection of the disease may result in the patient’s death. Thus, advanced diagnosis is a critical factor besides the disease progress. In addition to advanced diagnosis, the privacy of datasets is important for organizations. Due to the great value of datasets, hospitals do not want to share their datasets, but they want to share their trained network weights. Therefore, in this paper, we combine deep learning and blockchain to implement the blockchain as distributed storage. Using permission blockchain, weights are broadcasted among other hospitals securely. Because of the security, the dataset is shared with five hospitals equally. Each hospital trains its network model and sends its weights to the blockchain. The goal is to broadcast the aggregated weights among hospitals securely and have good enough results because the whole dataset is not implemented to train a network. The dataset contains 5856 images, and hospitals implement a residual neural network with 28 layers. The results show that hospitals can increase the accuracy of their model using shared weights compared to a model without using shared weights.","PeriodicalId":184716,"journal":{"name":"2022 International Conference on Machine Vision and Image Processing (MVIP)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123935381","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-02-23DOI: 10.1109/MVIP53647.2022.9738763
S. Aghapour Maleki, H. Ghassemian
Assessing the quality of pansharpened images is a critical issue in order to obtain a quantitative score to represent the quality and compare the performance of different fusion methods. Most of the introduced metrics for pansharpened image quality assessment, evaluate the spectral content of the image, while in different applications of remote sensing like detection and identification of image objects, spatial quality has an important role. In the current study, a new index for spatial quality assessment is introduced that extracts gray level co-occurrence matrix (GLCM) from distorted and reference images and compares the similarities of these features. The tempere image database 2013 (TID2013) that provides reference and different types of distorted images with subjective scores of each image is used as the desired database. To solve the high computational complexity of obtaining GLCM features, the fast GLCM method is employed. In this way, 16 different features are extracted. To select the features that have the most consistency with the human visual system (HVS), the forward floating search method is used as a feature selection method and five features are obtained as the final features to form the desired index. Experimental results show the efficiency of the proposed method in determining the spatial quality of fused images compared with that of the available quality assessment metrics.
{"title":"Spatial Quality Assessment of Pansharpened Images Based on Gray Level Co-Occurrence Matrix","authors":"S. Aghapour Maleki, H. Ghassemian","doi":"10.1109/MVIP53647.2022.9738763","DOIUrl":"https://doi.org/10.1109/MVIP53647.2022.9738763","url":null,"abstract":"Assessing the quality of pansharpened images is a critical issue in order to obtain a quantitative score to represent the quality and compare the performance of different fusion methods. Most of the introduced metrics for pansharpened image quality assessment, evaluate the spectral content of the image, while in different applications of remote sensing like detection and identification of image objects, spatial quality has an important role. In the current study, a new index for spatial quality assessment is introduced that extracts gray level co-occurrence matrix (GLCM) from distorted and reference images and compares the similarities of these features. The tempere image database 2013 (TID2013) that provides reference and different types of distorted images with subjective scores of each image is used as the desired database. To solve the high computational complexity of obtaining GLCM features, the fast GLCM method is employed. In this way, 16 different features are extracted. To select the features that have the most consistency with the human visual system (HVS), the forward floating search method is used as a feature selection method and five features are obtained as the final features to form the desired index. Experimental results show the efficiency of the proposed method in determining the spatial quality of fused images compared with that of the available quality assessment metrics.","PeriodicalId":184716,"journal":{"name":"2022 International Conference on Machine Vision and Image Processing (MVIP)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125272447","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-02-23DOI: 10.1109/MVIP53647.2022.9738555
Mostafa Amin-Naji, A. Aghagolzadeh, Hami Mahdavinataj
This paper presents fast pixel-wise multi-focus image fusion in the spatial domain without bells and whistles. The proposed method just uses the determinant of the sliding windows from the input images as a metric to create a pixel-wise decision map. The sliding windows of 15 pixels with the stride of 7 pixels are passed through the input images. Then it creates a pixel-wise decision map for fusion multi-focus images. Also, some simple tricks like global image threshold using Otsu’s method and removal of small objects by morphological closing operation are used to refine the pixel-wise decision map. This method is high-speed and can fuse a pair of 512x512 multi-focus images around 0.05 seconds (50 milliseconds) in our hardware. We compared it with 22 prominent methods in the transform domain, spatial domain, and deep learning based methods that their source codes are available, and our method is faster than all of them. We conducted the objective and subjective experiments on the Lytro dataset, and our method can compete with their results. The proposed method may not have the best fusion quality among state-of-the-art methods, but to the best of our knowledge, this is the fastest pixel-wise method and very suitable for real-time image processing. All material and source code will be available in https://github.com/mostafaaminnaji/FastDetFuse and http://imagefusion.ir.
{"title":"Fast Multi Focus Image Fusion Using Determinant","authors":"Mostafa Amin-Naji, A. Aghagolzadeh, Hami Mahdavinataj","doi":"10.1109/MVIP53647.2022.9738555","DOIUrl":"https://doi.org/10.1109/MVIP53647.2022.9738555","url":null,"abstract":"This paper presents fast pixel-wise multi-focus image fusion in the spatial domain without bells and whistles. The proposed method just uses the determinant of the sliding windows from the input images as a metric to create a pixel-wise decision map. The sliding windows of 15 pixels with the stride of 7 pixels are passed through the input images. Then it creates a pixel-wise decision map for fusion multi-focus images. Also, some simple tricks like global image threshold using Otsu’s method and removal of small objects by morphological closing operation are used to refine the pixel-wise decision map. This method is high-speed and can fuse a pair of 512x512 multi-focus images around 0.05 seconds (50 milliseconds) in our hardware. We compared it with 22 prominent methods in the transform domain, spatial domain, and deep learning based methods that their source codes are available, and our method is faster than all of them. We conducted the objective and subjective experiments on the Lytro dataset, and our method can compete with their results. The proposed method may not have the best fusion quality among state-of-the-art methods, but to the best of our knowledge, this is the fastest pixel-wise method and very suitable for real-time image processing. All material and source code will be available in https://github.com/mostafaaminnaji/FastDetFuse and http://imagefusion.ir.","PeriodicalId":184716,"journal":{"name":"2022 International Conference on Machine Vision and Image Processing (MVIP)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126981756","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Increasing attention to steganalysis and steganography due to the need for secure information transfer is one of the most important concerns of communication. Among the several image formats, JPEG is the most widely used compression method today. As a result, various stenographic systems based on disguising messages in jpeg format have been presented. Consequently, steganalysis of JPEG images is very essential. Recently, using neural networks and deep learning has greatly increased both in spatial and JPEG steganalysis. However, in the field of JPEG steganalysis, most of the existing networks still utilized hand-designed components as well. In the proposed JPEG steganalysis method we investigate the relations of the quantized Discrete Cosine Transform (DCT) coefficients and extract the binary vectors as the input of the neural network employing the relations of mid-frequency coefficients. The experimental results illustrate the acceptable detection rate of the simple presented approach.
{"title":"JPEG Steganalysis Using the Relations Between DCT Coefficients","authors":"Seyedeh Maryam Seyed Khalilollahi, Azadeh Mansouri","doi":"10.1109/MVIP53647.2022.9738785","DOIUrl":"https://doi.org/10.1109/MVIP53647.2022.9738785","url":null,"abstract":"Increasing attention to steganalysis and steganography due to the need for secure information transfer is one of the most important concerns of communication. Among the several image formats, JPEG is the most widely used compression method today. As a result, various stenographic systems based on disguising messages in jpeg format have been presented. Consequently, steganalysis of JPEG images is very essential. Recently, using neural networks and deep learning has greatly increased both in spatial and JPEG steganalysis. However, in the field of JPEG steganalysis, most of the existing networks still utilized hand-designed components as well. In the proposed JPEG steganalysis method we investigate the relations of the quantized Discrete Cosine Transform (DCT) coefficients and extract the binary vectors as the input of the neural network employing the relations of mid-frequency coefficients. The experimental results illustrate the acceptable detection rate of the simple presented approach.","PeriodicalId":184716,"journal":{"name":"2022 International Conference on Machine Vision and Image Processing (MVIP)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131400818","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：