Pub Date : 2022-12-15DOI: 10.1142/s0219467824500268
Demudu Naidu Chukka, J. Meka, S. Setty, P. Choppala
The focus of this paper is impulse noise reduction in digital color images. The most popular noise reduction schemes are the vector median filter and its many variants that operate by minimizing the aggregate distance from one pixel to every other pixel in a chosen window. This minimizing operation determines the most confirmative pixel based on its similarity to the chosen window and replaces the central pixel of the window with the determined one. The peer group filters, unlike the vector median filters, determine a set of pixels that are most confirmative to the window and then perform filtering over the determined set. Using a set of pixels in the filtering process rather than one pixel is more helpful as it takes into account the full information of all the pixels that seemingly contribute to the signal. Hence, the peer group filters are found to be more robust to noise. However, the peer group for each pixel is computed deterministically using thresholding schemes. A wrong choice of the threshold will easily impair the filtering performance. In this paper, we propose a peer group filtering approach using principles of Bayesian probability theory and clustering. Here, we present a method to compute the probability that a pixel value is clean (not corrupted by impulse noise) and then apply clustering on the probability measure to determine the peer group. The key benefit of this proposal is that the need for thresholding in peer group filtering is completely avoided. Simulation results show that the proposed method performs better than the conventional vector median and peer group filtering methods in terms of noise reduction and structural similarity, thus validating the proposed approach.
{"title":"Bayesian Selective Median Filtering for Reduction of Impulse Noise in Digital Color Images","authors":"Demudu Naidu Chukka, J. Meka, S. Setty, P. Choppala","doi":"10.1142/s0219467824500268","DOIUrl":"https://doi.org/10.1142/s0219467824500268","url":null,"abstract":"The focus of this paper is impulse noise reduction in digital color images. The most popular noise reduction schemes are the vector median filter and its many variants that operate by minimizing the aggregate distance from one pixel to every other pixel in a chosen window. This minimizing operation determines the most confirmative pixel based on its similarity to the chosen window and replaces the central pixel of the window with the determined one. The peer group filters, unlike the vector median filters, determine a set of pixels that are most confirmative to the window and then perform filtering over the determined set. Using a set of pixels in the filtering process rather than one pixel is more helpful as it takes into account the full information of all the pixels that seemingly contribute to the signal. Hence, the peer group filters are found to be more robust to noise. However, the peer group for each pixel is computed deterministically using thresholding schemes. A wrong choice of the threshold will easily impair the filtering performance. In this paper, we propose a peer group filtering approach using principles of Bayesian probability theory and clustering. Here, we present a method to compute the probability that a pixel value is clean (not corrupted by impulse noise) and then apply clustering on the probability measure to determine the peer group. The key benefit of this proposal is that the need for thresholding in peer group filtering is completely avoided. Simulation results show that the proposed method performs better than the conventional vector median and peer group filtering methods in terms of noise reduction and structural similarity, thus validating the proposed approach.","PeriodicalId":44688,"journal":{"name":"International Journal of Image and Graphics","volume":" ","pages":""},"PeriodicalIF":1.6,"publicationDate":"2022-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46527184","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 : 2022-12-15DOI: 10.1142/s0219467823400132
S. Praveen, V. E. Jyothi, Chokka Anuradha, K. VenuGopal, V. Shariff, S. Sindhura
Nowadays, in most countries, the most dangerous and life threatening infection is Chronic Kidney Disease (CKD). A progressive malfunctioning of the kidneys and less effectiveness of the kidney are considered CKD. CKD can be a life threatening disease if it continues for longer period of time. Prediction of chronic disease in early stage is very crucial so that sustainable care of the patient is taken to prevent menacing situations. Most of the developing countries are being affected by this deadly disease and treatment applied for this disease is also very expensive, here in this paper, a Machine Learning (ML)-positioned approach called Neuro-Fuzzy model is used for prediction belonging to CKD. Based on the image processing technique, fibrosis proportions are detected in the kidney tissues. It also builds a system for identifying and detection of CKD at an early stage. Neuro-Fuzzy model is based on ML which can detect risk of CKD patients. Compared with other conventional methods such as Support Vector Machine (SVM) and K-Nearest Neighbor (KNN), the proposed method of this paper — ML-based Neuro-Fuzzy logic method — obtained 97% accuracy in CKD prediction. This method can be evaluated based on various parameters such as Precision, Accuracy, Recall and F1-Score in CKD prediction. From the results, the patients having high risk of chronic disease can be predicted.
{"title":"Chronic Kidney Disease Prediction Using ML-Based Neuro-Fuzzy Model","authors":"S. Praveen, V. E. Jyothi, Chokka Anuradha, K. VenuGopal, V. Shariff, S. Sindhura","doi":"10.1142/s0219467823400132","DOIUrl":"https://doi.org/10.1142/s0219467823400132","url":null,"abstract":"Nowadays, in most countries, the most dangerous and life threatening infection is Chronic Kidney Disease (CKD). A progressive malfunctioning of the kidneys and less effectiveness of the kidney are considered CKD. CKD can be a life threatening disease if it continues for longer period of time. Prediction of chronic disease in early stage is very crucial so that sustainable care of the patient is taken to prevent menacing situations. Most of the developing countries are being affected by this deadly disease and treatment applied for this disease is also very expensive, here in this paper, a Machine Learning (ML)-positioned approach called Neuro-Fuzzy model is used for prediction belonging to CKD. Based on the image processing technique, fibrosis proportions are detected in the kidney tissues. It also builds a system for identifying and detection of CKD at an early stage. Neuro-Fuzzy model is based on ML which can detect risk of CKD patients. Compared with other conventional methods such as Support Vector Machine (SVM) and K-Nearest Neighbor (KNN), the proposed method of this paper — ML-based Neuro-Fuzzy logic method — obtained 97% accuracy in CKD prediction. This method can be evaluated based on various parameters such as Precision, Accuracy, Recall and F1-Score in CKD prediction. From the results, the patients having high risk of chronic disease can be predicted.","PeriodicalId":44688,"journal":{"name":"International Journal of Image and Graphics","volume":" ","pages":""},"PeriodicalIF":1.6,"publicationDate":"2022-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44430404","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 : 2022-12-14DOI: 10.1142/s0219467824500244
J. D. Pakhare, M. Uplane
Image classification in the image is the persistent task to be computed in robotics, automobiles, and machine vision for sustainability. Scene categorization remains one of the challenging parts of various multi-media technologies implied in human–computer communication, robotic navigation, video surveillance, medical diagnosing, tourist guidance, and drone targeting. In this research, a Hybrid Mayfly Lévy flight distribution (MLFD) optimization algorithm-tuned deep convolutional neural network is proposed to effectively classify the image. The feature extraction process is a significant task to be executed as it enhances the classifier performance by reducing the execution time and the computational complexity. Further, the classifier is optimally trained by the Hybrid MLFD algorithm which in turn reduces optimization issues. The accuracy of the proposed MLFD-based Deep-CNN using the SCID-2 dataset is 95.2683% at 80% of training and 97.6425% for 10 K-fold. This manifests that the proposed MLFD-based Deep-CNN outperforms all the conventional methods in terms of accuracy, sensitivity, and specificity.
{"title":"Hybrid Mayfly Lévy Flight Distribution Optimization Algorithm-Tuned Deep Convolutional Neural Network for Indoor–Outdoor Image Classification","authors":"J. D. Pakhare, M. Uplane","doi":"10.1142/s0219467824500244","DOIUrl":"https://doi.org/10.1142/s0219467824500244","url":null,"abstract":"Image classification in the image is the persistent task to be computed in robotics, automobiles, and machine vision for sustainability. Scene categorization remains one of the challenging parts of various multi-media technologies implied in human–computer communication, robotic navigation, video surveillance, medical diagnosing, tourist guidance, and drone targeting. In this research, a Hybrid Mayfly Lévy flight distribution (MLFD) optimization algorithm-tuned deep convolutional neural network is proposed to effectively classify the image. The feature extraction process is a significant task to be executed as it enhances the classifier performance by reducing the execution time and the computational complexity. Further, the classifier is optimally trained by the Hybrid MLFD algorithm which in turn reduces optimization issues. The accuracy of the proposed MLFD-based Deep-CNN using the SCID-2 dataset is 95.2683% at 80% of training and 97.6425% for 10 K-fold. This manifests that the proposed MLFD-based Deep-CNN outperforms all the conventional methods in terms of accuracy, sensitivity, and specificity.","PeriodicalId":44688,"journal":{"name":"International Journal of Image and Graphics","volume":" ","pages":""},"PeriodicalIF":1.6,"publicationDate":"2022-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44900484","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 : 2022-12-14DOI: 10.1142/s021946782450027x
Shuhan He, Xueming Li, Qiang Fu
Three-dimensional (3D) human mesh sequences obtained by 3D scanning equipment are often used in film and television, games, the internet, and other industries. However, due to the dense point cloud data obtained by 3D scanning equipment, the data of a single frame of a 3D human model is always large. Considering the different topologies of models between different frames, and even the interaction between the human body and other objects, the content of 3D models between different frames is also complex. Therefore, the traditional 3D model compression method always cannot handle the compression of the 3D human mesh sequence. To address this problem, we propose a sequence compression method of 3D human mesh sequence based on the Laplace operator, and test it on the complex interactive behavior of a soccer player bouncing the ball. This method first detects the mesh separation degree of the interactive object and human body, and then divides the sequence into a series of fragments based on the consistency of separation degrees. In each fragment, we employ a deformation algorithm to map keyframe topology to other frames, to improve the compression ratio of the sequence. Our work can be used for the storage of mesh sequences and mobile applications by providing an approach for data compression.
{"title":"Laplace-Based 3D Human Mesh Sequence Compression","authors":"Shuhan He, Xueming Li, Qiang Fu","doi":"10.1142/s021946782450027x","DOIUrl":"https://doi.org/10.1142/s021946782450027x","url":null,"abstract":"Three-dimensional (3D) human mesh sequences obtained by 3D scanning equipment are often used in film and television, games, the internet, and other industries. However, due to the dense point cloud data obtained by 3D scanning equipment, the data of a single frame of a 3D human model is always large. Considering the different topologies of models between different frames, and even the interaction between the human body and other objects, the content of 3D models between different frames is also complex. Therefore, the traditional 3D model compression method always cannot handle the compression of the 3D human mesh sequence. To address this problem, we propose a sequence compression method of 3D human mesh sequence based on the Laplace operator, and test it on the complex interactive behavior of a soccer player bouncing the ball. This method first detects the mesh separation degree of the interactive object and human body, and then divides the sequence into a series of fragments based on the consistency of separation degrees. In each fragment, we employ a deformation algorithm to map keyframe topology to other frames, to improve the compression ratio of the sequence. Our work can be used for the storage of mesh sequences and mobile applications by providing an approach for data compression.","PeriodicalId":44688,"journal":{"name":"International Journal of Image and Graphics","volume":" ","pages":""},"PeriodicalIF":1.6,"publicationDate":"2022-12-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42933950","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 : 2022-12-10DOI: 10.1142/s0219467824500220
Samina Mulla, N. Shaikh
In this digital era, there is a tremendous increase in the volume of data, which adds difficulties to the person who utilizes particular applications, such as websites, email, and news. Text summarization targets to reduce the complexity of obtaining statistics from the websites as it compresses the textual document to a short summary without affecting the relevant information. The crucial step in multi-document summarization is obtaining a relationship between the cross-sentence. However, the conventional methods fail to determine the inter-sentence relationship, especially in long documents. This research develops a graph-based neural network to attain an inter-sentence relationship. The significant step in the proposed multi-document text summarization model is forming the weighted graph embedding features. Furthermore, the weighted graph embedding features are utilized to evaluate the relationship between the document’s words and sentences. Finally, the bidirectional long short-term memory (BiLSTM) classifier is utilized to summarize the multi-document text summarization. The experimental analysis uses the three standard datasets, the Daily Mail dataset, Document Understanding Conference (DUC) 2002, and Document Understanding Conference (DUC) 2004 dataset. The experimental outcome demonstrates that the proposed weighted graph embedding feature + BiLSTM model exceeds all the conventional methods with Precision, Recall, and F1 score of 0.5352, 0.6296, and 0.5429, respectively.
{"title":"Weighted Graph Embedding Feature with Bi-Directional Long Short-Term Memory Classifier for Multi-Document Text Summarization","authors":"Samina Mulla, N. Shaikh","doi":"10.1142/s0219467824500220","DOIUrl":"https://doi.org/10.1142/s0219467824500220","url":null,"abstract":"In this digital era, there is a tremendous increase in the volume of data, which adds difficulties to the person who utilizes particular applications, such as websites, email, and news. Text summarization targets to reduce the complexity of obtaining statistics from the websites as it compresses the textual document to a short summary without affecting the relevant information. The crucial step in multi-document summarization is obtaining a relationship between the cross-sentence. However, the conventional methods fail to determine the inter-sentence relationship, especially in long documents. This research develops a graph-based neural network to attain an inter-sentence relationship. The significant step in the proposed multi-document text summarization model is forming the weighted graph embedding features. Furthermore, the weighted graph embedding features are utilized to evaluate the relationship between the document’s words and sentences. Finally, the bidirectional long short-term memory (BiLSTM) classifier is utilized to summarize the multi-document text summarization. The experimental analysis uses the three standard datasets, the Daily Mail dataset, Document Understanding Conference (DUC) 2002, and Document Understanding Conference (DUC) 2004 dataset. The experimental outcome demonstrates that the proposed weighted graph embedding feature + BiLSTM model exceeds all the conventional methods with Precision, Recall, and F1 score of 0.5352, 0.6296, and 0.5429, respectively.","PeriodicalId":44688,"journal":{"name":"International Journal of Image and Graphics","volume":"1 1","pages":""},"PeriodicalIF":1.6,"publicationDate":"2022-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41501572","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 : 2022-12-07DOI: 10.1142/s0219467824500177
M. N. Sowmya, K. Prasanna
Image retrieval is a significant and hot research topic among researchers that drives the focus of researchers from keyword toward semantic-based image reconstruction. Nevertheless, existing image retrieval investigations still have a shortage of significant semantic image definition and user behavior consideration. Hence, there is a necessity to offer a high level of assistance towards regulating the semantic gap between low-level visual patterns and high-level ideas for a better understanding between humans and machines. Hence, this research devises an effective medical image retrieval strategy using convoluted neighborhood-based Ordered-dither block truncation coding (ODBTC). The developed approach is devised by modifying the ODBTC concept using a convoluted neighborhood mechanism. Here, the convoluted neighborhood-based color co-occurrence feature (CCF) and convoluted neighborhood-based bit pattern feature (BBF) are extracted. Finally, cross-indexing is performed to convert the feature points into binary codes for effective image retrieval. Meanwhile, the proposed convoluted neighborhood-based ODBTC has achieved maximum precision, recall, and f-measure with values of 0.740, 0.680, and 0.709.
{"title":"Convoluted Neighborhood-Based Ordered-Dither Block Truncation Coding for Ear Image Retrieval","authors":"M. N. Sowmya, K. Prasanna","doi":"10.1142/s0219467824500177","DOIUrl":"https://doi.org/10.1142/s0219467824500177","url":null,"abstract":"Image retrieval is a significant and hot research topic among researchers that drives the focus of researchers from keyword toward semantic-based image reconstruction. Nevertheless, existing image retrieval investigations still have a shortage of significant semantic image definition and user behavior consideration. Hence, there is a necessity to offer a high level of assistance towards regulating the semantic gap between low-level visual patterns and high-level ideas for a better understanding between humans and machines. Hence, this research devises an effective medical image retrieval strategy using convoluted neighborhood-based Ordered-dither block truncation coding (ODBTC). The developed approach is devised by modifying the ODBTC concept using a convoluted neighborhood mechanism. Here, the convoluted neighborhood-based color co-occurrence feature (CCF) and convoluted neighborhood-based bit pattern feature (BBF) are extracted. Finally, cross-indexing is performed to convert the feature points into binary codes for effective image retrieval. Meanwhile, the proposed convoluted neighborhood-based ODBTC has achieved maximum precision, recall, and f-measure with values of 0.740, 0.680, and 0.709.","PeriodicalId":44688,"journal":{"name":"International Journal of Image and Graphics","volume":" ","pages":""},"PeriodicalIF":1.6,"publicationDate":"2022-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43886494","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 : 2022-12-05DOI: 10.1142/s0219467824500074
P. S. Subhashini Pedalanka, M. Satya Sai Ram, Duggirala Sreenivasa Rao
Automatic Speech Recognition (ASR) has been an intensive research area during the recent years in internet to enable natural human–machine communication. However, the existing Deep Neutral Network (DNN) techniques need more focus on feature extraction process and recognition accuracy. Thus, an enhanced deep neural network (DNN)-based approach for speaker recognition with a novel Triumvirate Euphemism Strategy (TES) is proposed. This overcomes poor feature extraction from Mel-Frequency Cepstral Coefficient (MFCC) map by extracting the features based on petite, hefty and artistry of the features. Then, the features are trained with Silhouette Martyrs Method (SMM) without any inter-class and intra-class separability problems and margins are affixed between classes with three new loss functions, namely A-Loss, AM-Loss and AAM-Loss. Additionally, the parallelization is done by a mini-batch-based BP algorithm in DNN. A novel Frenzied Heap Atrophy (FHA) with a multi-GPU model is introduced in addition with DNN to enhance the parallelized computing that accelerates the training procedures. Thus, the outcome of the proposed technique is highly efficient that provides feasible extraction features and gives incredibly precise results with 97.5% accuracy in the recognition of speakers. Moreover, various parameters were discussed to prove the efficiency of the system and also the proposed method outperformed the existing methods in all aspects.
{"title":"An Enhanced Deep Neural Network-Based Approach for Speaker Recognition Using Triumvirate Euphemism Strategy","authors":"P. S. Subhashini Pedalanka, M. Satya Sai Ram, Duggirala Sreenivasa Rao","doi":"10.1142/s0219467824500074","DOIUrl":"https://doi.org/10.1142/s0219467824500074","url":null,"abstract":"Automatic Speech Recognition (ASR) has been an intensive research area during the recent years in internet to enable natural human–machine communication. However, the existing Deep Neutral Network (DNN) techniques need more focus on feature extraction process and recognition accuracy. Thus, an enhanced deep neural network (DNN)-based approach for speaker recognition with a novel Triumvirate Euphemism Strategy (TES) is proposed. This overcomes poor feature extraction from Mel-Frequency Cepstral Coefficient (MFCC) map by extracting the features based on petite, hefty and artistry of the features. Then, the features are trained with Silhouette Martyrs Method (SMM) without any inter-class and intra-class separability problems and margins are affixed between classes with three new loss functions, namely A-Loss, AM-Loss and AAM-Loss. Additionally, the parallelization is done by a mini-batch-based BP algorithm in DNN. A novel Frenzied Heap Atrophy (FHA) with a multi-GPU model is introduced in addition with DNN to enhance the parallelized computing that accelerates the training procedures. Thus, the outcome of the proposed technique is highly efficient that provides feasible extraction features and gives incredibly precise results with 97.5% accuracy in the recognition of speakers. Moreover, various parameters were discussed to prove the efficiency of the system and also the proposed method outperformed the existing methods in all aspects.","PeriodicalId":44688,"journal":{"name":"International Journal of Image and Graphics","volume":" ","pages":""},"PeriodicalIF":1.6,"publicationDate":"2022-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49309807","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 : 2022-11-24DOI: 10.1142/s021946782350047x
S. Jameel, Jafar Majidpour
For Compressive Sensing problems, a number of techniques have been introduced, including traditional compressed-sensing (CS) image reconstruction and Deep Neural Network (DNN) models. Unfortunately, due to low sampling rates, the quality of image reconstruction is still poor. This paper proposes a lossy image compression model (i.e. BCS-AE), which combines two different types to produce a model that uses more high-quality low-bitrate CS reconstruction. Initially, block-based compressed sensing (BCS) was utilized, and it was done one block at a time by the same operator. It can correctly extract images with complex geometric configurations. Second, we create an AutoEncoder architecture to replace traditional transforms, and we train it with a rate-distortion loss function. The proposed model is trained and then tested on the CelebA and Kodak databases. According to the results, advanced deep learning-based and iterative optimization-based algorithms perform better in terms of compression ratio and reconstruction quality.
{"title":"BCS-AE: Integrated Image Compression-Encryption Model Based on AE and Block-CS","authors":"S. Jameel, Jafar Majidpour","doi":"10.1142/s021946782350047x","DOIUrl":"https://doi.org/10.1142/s021946782350047x","url":null,"abstract":"For Compressive Sensing problems, a number of techniques have been introduced, including traditional compressed-sensing (CS) image reconstruction and Deep Neural Network (DNN) models. Unfortunately, due to low sampling rates, the quality of image reconstruction is still poor. This paper proposes a lossy image compression model (i.e. BCS-AE), which combines two different types to produce a model that uses more high-quality low-bitrate CS reconstruction. Initially, block-based compressed sensing (BCS) was utilized, and it was done one block at a time by the same operator. It can correctly extract images with complex geometric configurations. Second, we create an AutoEncoder architecture to replace traditional transforms, and we train it with a rate-distortion loss function. The proposed model is trained and then tested on the CelebA and Kodak databases. According to the results, advanced deep learning-based and iterative optimization-based algorithms perform better in terms of compression ratio and reconstruction quality.","PeriodicalId":44688,"journal":{"name":"International Journal of Image and Graphics","volume":" ","pages":""},"PeriodicalIF":1.6,"publicationDate":"2022-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41789443","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 : 2022-11-24DOI: 10.1142/s0219467824500207
Fuhe Du, B. Peng, Zaid Al-Huda, Jing Yao
Deep learning-based skin lesion segmentation methods have achieved promising results in the community. However, they are usually based on fully supervised learning and require many high-quality ground truths. Labeling the ground truths takes a lot of labor, material, and financial resources. We propose a novel semi-supervised skin lesion segmentation method to solve this problem. First, a hierarchical image segmentation algorithm is used to generate optimal segmentation maps. Then, fully supervised training is performed on a small part of the images with ground truths. The resulting pseudo masks are generated to train the rest of the images. The optimal segmentation maps are utilized in this process to refine the pseudo masks. Experiments show that the proposed method can improve the performance of semi-supervised learning for skin lesion segmentation by reducing the gap with fully supervised learning methods. Moreover, it can reduce the workload of labeling the ground truths. Extensive experiments are conducted on the open dataset to validate the efficiency of the proposed method. The results show that our method is competitive in improving the quality of semi-supervised segmentation.
{"title":"Semi-Supervised Skin Lesion Segmentation via Iterative Mask Optimization","authors":"Fuhe Du, B. Peng, Zaid Al-Huda, Jing Yao","doi":"10.1142/s0219467824500207","DOIUrl":"https://doi.org/10.1142/s0219467824500207","url":null,"abstract":"Deep learning-based skin lesion segmentation methods have achieved promising results in the community. However, they are usually based on fully supervised learning and require many high-quality ground truths. Labeling the ground truths takes a lot of labor, material, and financial resources. We propose a novel semi-supervised skin lesion segmentation method to solve this problem. First, a hierarchical image segmentation algorithm is used to generate optimal segmentation maps. Then, fully supervised training is performed on a small part of the images with ground truths. The resulting pseudo masks are generated to train the rest of the images. The optimal segmentation maps are utilized in this process to refine the pseudo masks. Experiments show that the proposed method can improve the performance of semi-supervised learning for skin lesion segmentation by reducing the gap with fully supervised learning methods. Moreover, it can reduce the workload of labeling the ground truths. Extensive experiments are conducted on the open dataset to validate the efficiency of the proposed method. The results show that our method is competitive in improving the quality of semi-supervised segmentation.","PeriodicalId":44688,"journal":{"name":"International Journal of Image and Graphics","volume":" ","pages":""},"PeriodicalIF":1.6,"publicationDate":"2022-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46872995","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 : 2022-11-18DOI: 10.1142/s0219467823400065
S. Gadde, K. Kiran
Diabetic retinopathy (DR) is stated as a disease in the eyes that affects the retina blood vessels and causes blindness. The early diagnosis and detection of the DR in patients preserve the patient’s vision. In general, for the diagnosis of eye diseases, retinal fundus images are employed. The advancement in the automatic diagnosis of diseases attained higher significance for rapid advancement in computing technology in the medical field. Besides, for the diagnosis of the diseases, fundus image automatic detection is involved in the recognition of blood vessels evaluated based on the length, branching pattern, and width. However, fundus images have low contrast and it is difficult to evaluate the identification of the disease in blood vessels. As a result, it is necessary to adopt a consistent automated method to extract blood vessels in the fundus images for DR. The conventional automated localization of the macula and optic disk in the retinal fundus images needs to be improved for DR disease diagnosis. But existing methods are not sufficient for the early identification and detection of DR. This paper proposed an entropy distributed matching global and local clustering (EDMGL) for fundus images. The developed EDMGL comprises the different uncertainties for the evaluation of the classes based on local and global entropy. The fundus image local entropy is evaluated based on the spatial likelihood fuzzifier membership function estimation for segmentation. The final proposed algorithm membership function is estimated using the addition of weighted parameters through membership estimation based on the global and local entropy. The classification performance of the proposed EDMGL is evaluated based on the dice coefficient, segmentation accuracy, and partition entropy. The performance of the proposed EDMGL is comparatively examined with the conventional technique. The comparative analysis expressed that the performance of the proposed EDMGL exhibits [Formula: see text]5% improved performance in terms of accuracy, precision, recall, and F1-score.
{"title":"Entropy-Based Feature Extraction Model for Fundus Images with Deep Learning Model","authors":"S. Gadde, K. Kiran","doi":"10.1142/s0219467823400065","DOIUrl":"https://doi.org/10.1142/s0219467823400065","url":null,"abstract":"Diabetic retinopathy (DR) is stated as a disease in the eyes that affects the retina blood vessels and causes blindness. The early diagnosis and detection of the DR in patients preserve the patient’s vision. In general, for the diagnosis of eye diseases, retinal fundus images are employed. The advancement in the automatic diagnosis of diseases attained higher significance for rapid advancement in computing technology in the medical field. Besides, for the diagnosis of the diseases, fundus image automatic detection is involved in the recognition of blood vessels evaluated based on the length, branching pattern, and width. However, fundus images have low contrast and it is difficult to evaluate the identification of the disease in blood vessels. As a result, it is necessary to adopt a consistent automated method to extract blood vessels in the fundus images for DR. The conventional automated localization of the macula and optic disk in the retinal fundus images needs to be improved for DR disease diagnosis. But existing methods are not sufficient for the early identification and detection of DR. This paper proposed an entropy distributed matching global and local clustering (EDMGL) for fundus images. The developed EDMGL comprises the different uncertainties for the evaluation of the classes based on local and global entropy. The fundus image local entropy is evaluated based on the spatial likelihood fuzzifier membership function estimation for segmentation. The final proposed algorithm membership function is estimated using the addition of weighted parameters through membership estimation based on the global and local entropy. The classification performance of the proposed EDMGL is evaluated based on the dice coefficient, segmentation accuracy, and partition entropy. The performance of the proposed EDMGL is comparatively examined with the conventional technique. The comparative analysis expressed that the performance of the proposed EDMGL exhibits [Formula: see text]5% improved performance in terms of accuracy, precision, recall, and F1-score.","PeriodicalId":44688,"journal":{"name":"International Journal of Image and Graphics","volume":" ","pages":""},"PeriodicalIF":1.6,"publicationDate":"2022-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43949569","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：