{"title":"Enhanced Material Classification via MobileSEMNet: Leveraging MobileNetV2 for SEM Image Analysis","authors":"Cihat Aydin","doi":"10.18280/ts.400638","DOIUrl":"https://doi.org/10.18280/ts.400638","url":null,"abstract":"","PeriodicalId":49430,"journal":{"name":"Traitement Du Signal","volume":null,"pages":null},"PeriodicalIF":1.9,"publicationDate":"2023-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139139096","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Personalized Learning Pathway Generation for Online Education Through Image Recognition","authors":"Jie Yan, Na Wang, Yiming Wei, Menglu Han","doi":"10.18280/ts.400640","DOIUrl":"https://doi.org/10.18280/ts.400640","url":null,"abstract":"","PeriodicalId":49430,"journal":{"name":"Traitement Du Signal","volume":null,"pages":null},"PeriodicalIF":1.9,"publicationDate":"2023-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139139448","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Detection and Classification of Plant Stress Using Hybrid Deep Convolution Neural Networks: A Multi-Scale Vision Transformer Approach","authors":"Bhargavi Thokala, Sumathi Doraikannan","doi":"10.18280/ts.400625","DOIUrl":"https://doi.org/10.18280/ts.400625","url":null,"abstract":"","PeriodicalId":49430,"journal":{"name":"Traitement Du Signal","volume":null,"pages":null},"PeriodicalIF":1.9,"publicationDate":"2023-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139139571","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Leveraging Deep Learning for Identification of Illicit Images in Digital Forensic Investigations","authors":"M. Eriş, Mustafa Kaya","doi":"10.18280/ts.400617","DOIUrl":"https://doi.org/10.18280/ts.400617","url":null,"abstract":"","PeriodicalId":49430,"journal":{"name":"Traitement Du Signal","volume":null,"pages":null},"PeriodicalIF":1.9,"publicationDate":"2023-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139140673","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Optimizing Water Body Detection in Southeast Hubei Using PCA on Landsat ETM+ Imagery","authors":"Xiaohong Xiao, Jiangang Xie, Bin Cai","doi":"10.18280/ts.400612","DOIUrl":"https://doi.org/10.18280/ts.400612","url":null,"abstract":"","PeriodicalId":49430,"journal":{"name":"Traitement Du Signal","volume":null,"pages":null},"PeriodicalIF":1.9,"publicationDate":"2023-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139141852","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In the realm of face recognition utilising Support Vector Machines (SVM), the adaptivity of the penalty parameter c and the kernel function g is often found lacking, leading to suboptimal recognition rates. To address this issue, an approach harnessing the Sparrow Search Algorithm (SSA) for SVM parameter optimisation has been proposed. Traditional methods such as grid and random search, alongside other swarm intelligence optimisation algorithms like Particle Swarm Algorithm (PSO) and Differential Evolutionary Algorithm (DE), were surpassed by the capabilities of the SSA in numerous applications. Cross-validation (CV) was employed, with the SVM model training recognition accuracy serving as the SSA fitness value. Upon achieving optimal fitness values, the best combination of hyperparameters was ascertained. The overarching aim was to deploy the SSA for global optimisation of SVM's penalty parameters and kernel function, ensuring the derivation of the globally optimal solution for the ultimate classifier model in face recognition tasks. An empirical analysis conducted on the ORL standard face database revealed that the proposed method outperformed the PSO, DE, Gray Wolf Algorithm (GWO), and Enhanced Gray Wolf Algorithm (EGWO), registering an average accuracy of 95.1%. This starkly contrasts with the 81.9% accuracy of the traditional SVM. Such results demonstrate the method's efficacy in enhancing recognition performance, offering a novel avenue to elevate the accuracy of conventional SVM-based face recognition.
{"title":"Support Vector Machines Optimisation for Face Recognition Using Sparrow Search Algorithm","authors":"Wenli Lei, Yang Lei, Bin Li, Kun Jia","doi":"10.18280/ts.400519","DOIUrl":"https://doi.org/10.18280/ts.400519","url":null,"abstract":"In the realm of face recognition utilising Support Vector Machines (SVM), the adaptivity of the penalty parameter c and the kernel function g is often found lacking, leading to suboptimal recognition rates. To address this issue, an approach harnessing the Sparrow Search Algorithm (SSA) for SVM parameter optimisation has been proposed. Traditional methods such as grid and random search, alongside other swarm intelligence optimisation algorithms like Particle Swarm Algorithm (PSO) and Differential Evolutionary Algorithm (DE), were surpassed by the capabilities of the SSA in numerous applications. Cross-validation (CV) was employed, with the SVM model training recognition accuracy serving as the SSA fitness value. Upon achieving optimal fitness values, the best combination of hyperparameters was ascertained. The overarching aim was to deploy the SSA for global optimisation of SVM's penalty parameters and kernel function, ensuring the derivation of the globally optimal solution for the ultimate classifier model in face recognition tasks. An empirical analysis conducted on the ORL standard face database revealed that the proposed method outperformed the PSO, DE, Gray Wolf Algorithm (GWO), and Enhanced Gray Wolf Algorithm (EGWO), registering an average accuracy of 95.1%. This starkly contrasts with the 81.9% accuracy of the traditional SVM. Such results demonstrate the method's efficacy in enhancing recognition performance, offering a novel avenue to elevate the accuracy of conventional SVM-based face recognition.","PeriodicalId":49430,"journal":{"name":"Traitement Du Signal","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136067542","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Rapid advancements in artificial intelligence, machine learning, deep learning, coupled with easy access to high-capacity processing hardware, expansive organized datasets, and the evolution of artificial intelligence algorithms, have extensively influenced numerous fields. Digital Forensics is one such discipline where the application of artificial intelligence has been significantly amplified in recent years. The analysis of extensive image and video files derived from forensic evidence presents challenges in terms of time efficiency and accuracy. To surmount these challenges, artificial intelligence models can be employed to perform identification and classification processes on these data, thus expediting the resolution of forensic cases with enhanced precision. In the current study, state-of-the-art pre-trained YOLOv8 object recognition models - nano, small, medium, large, and extra-large - were utilized. These models were trained on the Wider-Face dataset with the objective of identifying suspects from images and videos sourced from digital materials in the field of digital forensics. The models achieved mean Average Precision (mAP) values of 97.513%, 98.569%, 98.763%, 98.775%, and 99.032% respectively. The YOLOv8 architecture demonstrated superior performance, outperforming the YOLOv5 architecture by a margin of 7.1% to 8.8%. To aid digital forensic experts in the detection and identification of suspicious individuals, a desktop application capable of real-time image analysis was developed.
{"title":"Real-Time Detection and Identification of Suspects in Forensic Imagery Using Advanced YOLOv8 Object Recognition Models","authors":"Serkan Karakuş, Mustafa Kaya, Seda Arslan Tuncer","doi":"10.18280/ts.400521","DOIUrl":"https://doi.org/10.18280/ts.400521","url":null,"abstract":"Rapid advancements in artificial intelligence, machine learning, deep learning, coupled with easy access to high-capacity processing hardware, expansive organized datasets, and the evolution of artificial intelligence algorithms, have extensively influenced numerous fields. Digital Forensics is one such discipline where the application of artificial intelligence has been significantly amplified in recent years. The analysis of extensive image and video files derived from forensic evidence presents challenges in terms of time efficiency and accuracy. To surmount these challenges, artificial intelligence models can be employed to perform identification and classification processes on these data, thus expediting the resolution of forensic cases with enhanced precision. In the current study, state-of-the-art pre-trained YOLOv8 object recognition models - nano, small, medium, large, and extra-large - were utilized. These models were trained on the Wider-Face dataset with the objective of identifying suspects from images and videos sourced from digital materials in the field of digital forensics. The models achieved mean Average Precision (mAP) values of 97.513%, 98.569%, 98.763%, 98.775%, and 99.032% respectively. The YOLOv8 architecture demonstrated superior performance, outperforming the YOLOv5 architecture by a margin of 7.1% to 8.8%. To aid digital forensic experts in the detection and identification of suspicious individuals, a desktop application capable of real-time image analysis was developed.","PeriodicalId":49430,"journal":{"name":"Traitement Du Signal","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136067781","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In many educational institutions, safety management traditionally depends upon manual video surveillance, leading to potential delays in the identification and alerting of perilous activities, notably the possession of controlled knives and smoking behaviors exhibited by students. These activities possess significant consequences for both the psychological and physical well-being of students. Recognizing this pressing need, an augmented object detection method for campus security, rooted in YOLOv7, is presented. The EIoU (Efficient Intersection over Union) loss function has been substituted to expedite model convergence and heighten detection fidelity. Additionally, the integration of the CBAM (Convolutional Block Attention Module) attention mechanism with the DCNv2 (Deformable ConvNets v2) deformable convolutional kernel not only mitigates the challenge of information inundation but also enhances feature extraction capabilities, facilitating adjustments to geometric deformations. Experimental findings indicate that this proposed method achieves a detection accuracy of 92.6% across various categories on a dataset comprising three categories, spanning a total of 4500 images, and attains an mAP of 96.4%. In comparison to the conventional YOLOv7 algorithm, enhancements in detection accuracy and mAP by 6.9% and 6.6%, respectively, have been observed, affirming the efficacy of the presented algorithm.
{"title":"Enhanced Campus Security Target Detection Using a Refined YOLOv7 Approach","authors":"Fengyun Cao, Shuai Ma","doi":"10.18280/ts.400544","DOIUrl":"https://doi.org/10.18280/ts.400544","url":null,"abstract":"In many educational institutions, safety management traditionally depends upon manual video surveillance, leading to potential delays in the identification and alerting of perilous activities, notably the possession of controlled knives and smoking behaviors exhibited by students. These activities possess significant consequences for both the psychological and physical well-being of students. Recognizing this pressing need, an augmented object detection method for campus security, rooted in YOLOv7, is presented. The EIoU (Efficient Intersection over Union) loss function has been substituted to expedite model convergence and heighten detection fidelity. Additionally, the integration of the CBAM (Convolutional Block Attention Module) attention mechanism with the DCNv2 (Deformable ConvNets v2) deformable convolutional kernel not only mitigates the challenge of information inundation but also enhances feature extraction capabilities, facilitating adjustments to geometric deformations. Experimental findings indicate that this proposed method achieves a detection accuracy of 92.6% across various categories on a dataset comprising three categories, spanning a total of 4500 images, and attains an mAP of 96.4%. In comparison to the conventional YOLOv7 algorithm, enhancements in detection accuracy and mAP by 6.9% and 6.6%, respectively, have been observed, affirming the efficacy of the presented algorithm.","PeriodicalId":49430,"journal":{"name":"Traitement Du Signal","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136068294","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Fahad M. Alshagathrh, Saleh Musleh, Mahmood Alzubaidi, Jens Schneider, Mowafa S. Househ
ABSTRACT
{"title":"Efficient Detection of Hepatic Steatosis in Ultrasound Images Using Convolutional Neural Networks: A Comparative Study","authors":"Fahad M. Alshagathrh, Saleh Musleh, Mahmood Alzubaidi, Jens Schneider, Mowafa S. Househ","doi":"10.18280/ts.400501","DOIUrl":"https://doi.org/10.18280/ts.400501","url":null,"abstract":"ABSTRACT","PeriodicalId":49430,"journal":{"name":"Traitement Du Signal","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136102825","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
As color images have been widely used in many fields, their restoration problem has received wide attention from researchers. This study proposed two solutions for denoising and low illuminance enhancement problems of existing color image restoration methods. At first, this paper built a colour image denoising model of weighted Schatten-p norm based on deep learning, which fully considers differences in the noise level of each channel of colour images, and could give a better denoising effect. Then, this study proposed a low illuminance color image enhancement algorithm that combines Gamma transform and Contrast Limited Adaptive Histogram Equalization (CLAHE), which could better balance image contrast enhancement and noise suppression. Studies of these two parts have both gained good results in terms of theory and experiment, and they could push the progress of colour image restoration technology and provide valuable references for related fields.
{"title":"A New Deep Learning-Based Restoration Method for Colour Images","authors":"Songshan Zu","doi":"10.18280/ts.400536","DOIUrl":"https://doi.org/10.18280/ts.400536","url":null,"abstract":"As color images have been widely used in many fields, their restoration problem has received wide attention from researchers. This study proposed two solutions for denoising and low illuminance enhancement problems of existing color image restoration methods. At first, this paper built a colour image denoising model of weighted Schatten-p norm based on deep learning, which fully considers differences in the noise level of each channel of colour images, and could give a better denoising effect. Then, this study proposed a low illuminance color image enhancement algorithm that combines Gamma transform and Contrast Limited Adaptive Histogram Equalization (CLAHE), which could better balance image contrast enhancement and noise suppression. Studies of these two parts have both gained good results in terms of theory and experiment, and they could push the progress of colour image restoration technology and provide valuable references for related fields.","PeriodicalId":49430,"journal":{"name":"Traitement Du Signal","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136103190","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}