检索结果-内蒙古大学图书馆

Re-identification in video surveillance systems considering appearance changes

Scientia Iranica 2024年第2期31卷 103-117页

作者： Mortezaie, Z. Hassanpour, H. Beghdadi, A. Faculty of Computer Engineering Shahrood University of Technology Shahrood Iran Institut Galilée Université Sorbonne Paris Nord Villetaneuse France

Human behavior analysis and visual anomaly detection are important applications in fields such as video surveillance, security systems, intelligent houses, and elderly care. People re-identification is one of the main steps in a surveillance system that directly affects system performance;and variations in appearance, pose, and scene illumination may be challenging issues for such a system. Previous re-identification approaches faced limitations while considering appearance changes in their tracking task. This paper proposes a new approach for people’s re-identification using a descriptor that is robust to appearance changes. In our proposed method, the enhanced Gaussian Of Gaussian (GOG) and the Hierarchical Gaussian Descriptors (HGDs) are employed to extract feature vectors from images. Experimental results on a number of commonly used people re-identification databases imply the superiority of the proposed approach in people re-identification compared to other existing approaches. © 2024 Sharif University of technology.

关键词： Security systems

来源：评论

学校读者我要写书评

暂无评论

An Investigation on Open-RAN Specifications:Use Cases,Security Threats,Requirements,Discussions

引用

computer Modeling in engineering & Sciences 2024年第10期141卷 13-41页

作者： Heejae Park Tri-HaiNguyen Laihyuk Park Department of Computer Science and Engineering Seoul National University of Science and TechnologySeoul01811Republic of Korea Faculty of Information Technology School of TechnologyVan Lang UniversityHo Chi Minh City70000Vietnam

The emergence of various technologies such as terahertz communications,Reconfigurable Intelligent Surfaces(RIS),and AI-powered communication services will burden network operators with rising infrastructure ***,the Open Radio Access Network(O-RAN)has been introduced as a solution for growing financial and operational burdens in Beyond 5G(B5G)and 6G networks.O-RAN promotes openness and intelligence to overcome the limitations of traditional *** disaggregating conventional Base Band Units(BBUs)into O-RAN Distributed Units(O-DU)and O-RAN Centralized Units(O-CU),O-RAN offers greater flexibility for upgrades and network ***,this openness introduces new security challenges compared to traditional *** existing studies overlook these security requirements of the O-RAN *** gain deeper insights into the O-RAN system and security,this paper first provides an overview of the general O-RAN architecture and its diverse use cases relevant to B5G and 6G *** then delve into specifications of O-RAN security threats and requirements,aiming to mitigate security vulnerabilities *** providing a comprehensive understanding of O-RAN architecture,use cases,and security considerations,thisworkserves as a valuable resource for future research in O-RAN and its security.

关键词： O-RAN architecture use cases security issues security requirements security discussions

来源：评论

学校读者我要写书评

暂无评论

An Efficient Encryption and Compression of Sensed IoT Medical Images Using Auto-Encoder

引用

computer Modeling in engineering & Sciences 2023年第2期134卷 909-926页

作者： Passent El-kafrawy Maie Aboghazalah Abdelmoty M.Ahmed Hanaa Torkey Ayman El-Sayed School of Information Technology and Computer Science Nile UniversityGiza12677Egypt Math and Computer Science Department Faculty of ScienceMenoufia UniversityMenoufia32511Egypt Department of Computer Engineering College of Computer ScienceKing Khalid UniversityAbha61421Saudi Arabia Computer Science and Engineering Department Faculty of Electronic EngineeringMenoufia UniversityMenouf32952Egypt

Healthcare systems nowadays depend on IoT sensors for sending data over the internet as a common *** ofmedical images is very important to secure patient *** these images consumes a lot of time onedge computing;therefore,theuse of anauto-encoder for compressionbefore encodingwill solve such a *** this paper,we use an auto-encoder to compress amedical image before encryption,and an encryption output(vector)is sent out over the *** the other hand,a decoder was used to reproduce the original image back after the vector was received and *** convolutional neural networks were conducted to evaluate our proposed approach:The first one is the auto-encoder,which is utilized to compress and encrypt the images,and the other assesses the classification accuracy of the image after decryption and *** hyperparameters of the encoder were tested,followed by the classification of the image to verify that no critical information was lost,to test the encryption and encoding *** this approach,sixteen hyperparameter permutations are utilized,but this research discusses three main cases in *** first case shows that the combination of Mean Square Logarithmic Error(MSLE),ADAgrad,two layers for the auto-encoder,and ReLU had the best auto-encoder results with a Mean Absolute Error(MAE)=0.221 after 50 epochs and 75%classification with the best result for the classification *** second case shows the reflection of auto-encoder results on the classification results which is a combination ofMean Square Error(MSE),RMSprop,three layers for the auto-encoder,and ReLU,which had the best classification accuracy of 65%,the auto-encoder gives MAE=0.31 after 50 *** third case is the worst,which is the combination of the hinge,RMSprop,three layers for the auto-encoder,and ReLU,providing accuracy of 20%and MAE=0.485.

关键词： Auto-encoder cloud image encryption IoT healthcare

来源：评论

学校读者我要写书评

暂无评论

A new method for promoting the detection of breast cancer in thermograms: applying deep autoencoders for eliminating redundancies in parallel with preserving independent components

引用

Journal of Ambient Intelligence and Humanized Computing 2024年第12期15卷 4085-4099页

作者： Shojaedini, Seyed Vahab Bahramzadeh, Bahram Biomedical Engineering Department Iranian Research Organization for Science & ampTechnology Tehran Iran Department of Computer Engineering Faculty of Engineering Islamic Azad University E-Campus Tehran Iran

Although thermography has been proposed over the past decade as an effective method for breast cancer diagnosis, the complexity of thermograms presents a significant obstacle, making their interpretation challenging. Recently, deep learning has emerged as a promising framework for analyzing breast thermograms. However, the performance of deep neural networks heavily relies on the volume of training images, and the limited number of labeled thermograms necessitates the generation of synthetic images for training. This article introduces a novel method that eliminates redundant or destructive information in synthetic thermograms while preserving essential and independent features using deep autoencoder concepts. Consequently, the proposed strategy enhances the representation of synthetic images for training deep networks, leading to improved breast cancer detection in thermograms. Evaluating the performance of the proposed method on the DMR-IR dataset shows a significant improvement in detecting cancerous breasts in thermograms compared to benchmark methods. Integrating this innovation into the basic model achieved an average accuracy of 92.3%, sensitivity of 93%, and specificity of 91.4%, surpassing the basic model’s 89.1%, 86%, and 92.5%, respectively. The proposed method demonstrated superior performance in mitigating overfitting, as evidenced by a smaller gap between training and validation curves, resulting in a 7% improvement in accuracy and a 3.2% increase in sensitivity. Although there was a slight decrease of 1.1% in specificity, it was outweighed by other parameter improvements. Additionally, the method led to a significant reduction in the dispersion range of test results, approximately six times less than alternative approaches. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2024.

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

Optimized digital watermarking: Harnessing the synergies of Schur matrix factorization, DCT, and DWT for superior image ownership proofing

引用

Multimedia Tools and Applications 2025年第18期84卷 19817-19852页

作者： Al-Aiash, Issa Alquran, Rabee AlJamal, Mahmoud Alsarhan, Ayoub Aljaidi, Mohammad Al-Fraihat, Dimah Department of Information Technology Faculty of Prince Al-Hussein Bin Abdallah II for Information Technology The Hashemite University Zarqa13133 Jordan Department of Computer Science Faculty of Information Technology Zarqa University Zarqa13110 Jordan Department of Software Engineering Faculty of Information Technology Isra University Amman11622 Jordan

The contemporary context of abundant digital dissemination inherently gives rise to the need for media protection and the clear identification of ownership rights. This paper responded to this nagging issue of unauthorized media capture by developing a new digital watermarking approach uniting the Discrete Wavelet Transform DWT, Discrete Cosine Transform DCT, and Schur matrix factorization. The proposed watermark, which was a positive vector uniformly distributed, presented a high degree of robustness to several adversarial operating conditions, including histogram equalization, salt and pepper noise, ripple impact, smoothing attack, and cropping. The obtained results reveal the effectiveness of our approach. The correlation coefficient between watermarked and original images remains exceptionally high at 0.999998, ensuring minimal visual distortion. Mean Squared Error (MSE) graphs for both high-frequency (HH) and low-frequency (LL) domains demonstrate negligible changes after attacks, confirming the watermark's robustness. Notably, the watermark extraction process remains successful even after histogram equalization, as indicated by significantly lower MSE compared to random vectors. Moreover, the average processing times for different attacks are promising;Histogram Equalization and Salt and Pepper attacks each took 0.3 s, Ripple Attack required 0.28 s, Smoothing Attack completed in 0.29 s, and the Cropping Attack took 0.6 s. The robustness of our approach appears against the salt and pepper attack, which historically presented a vulnerability when employing Schur matrix factorization in image watermarking, indicating the novelty of and this contribution to improving media ownership verification. By combining disparate transformation methodologies, this technique offers a promising avenue for reinforcing the integrity and authenticity of digital media, thereby strengthening the foundations of intellectual property rights in the digital domain. © The Author(s), und

关键词： Discrete wavelet transforms

来源：评论

学校读者我要写书评

暂无评论

In-service performance of emergency shutdown valves and dependent operational relationships in the offshore oil and gas industry

引用

Petroleum 2023年第4期9卷 613-620页

作者： Keith Cameron Andrew Lewis Diogo Montalvao Mohammad Reza Herfatmanesh School of Physics Engineering and Computer ScienceUniversity of HertfordshireUnited Kingdom Department of Design and Engineering Faculty of Science and TechnologyBournemouth UniversityUnited Kingdom

Industrial process plants use emergency shutdown valves(ESDVs)as safety barriers to protect against hazardous events,bringing the plant to a safe state when potential danger is *** ESDVs are used extensively in offshore oil and gas processing plants and have been mandated in the design of such systems from national and international standards and *** paper has used actual ESDV operating data from four mid/late life oil and gas production platforms in the North Sea to research operational relationships that are of interest to those responsible for the technical management and operation of *** first of the two relationships is between the closure time(CT)of the ESDV and the time it remains in the open position,prior to the close *** has been hypothesised that the CT of the ESDV is affected by the length of time that it has been open prior to being closed(Time since the last stroke).In addition to the general analysis of the data series,two sub-categories were created to further investigate this possible relationship for CT and these are“above mean”and“below mean”.The correlations(Pearson's based)resulting from this analysis are in the“weak”and“very weak”*** second relationship investigated was the effect of very frequent closures to assess if this improves the *** operational records for six subjects were analysed to find closures that occurred within a 24 h period of each ***,no discriminating trend was apparent where CT was impacted positively or negatively by the frequent closure *** was concluded that the variance of ESDV closure time cannot be influenced by the technical management of the ESDV in terms of scheduling the operation of the ESDV.

关键词： Emergency Shutdown Valve Safety instrumented systems Closure time IEC-61511

来源：评论

学校读者我要写书评

暂无评论

iLIAC: An approach of identifying dissimilar groups on unstructured numerical image dataset using improved agglomerative clustering technique

引用

Multimedia Tools and Applications 2024年第39期83卷 86359-86381页

作者： S, Sreedhar Kumar Ahmed, Syed Thouheed Fathima, Afifa Salsabil Mathivanan, Sandeep Kumar Jayagopal, Prabhu Saif, Abdu Gupta, Sachin Kumar Sinha, Garima Department of Computer Science and Engineering School of Engineering and Technology CMR University Karnataka Bengaluru India Department of Electrical Engineering Indian Institute of Technology Hyderabad Telangana Kandi India School of Computer Science and Engineering REVA University Bangalore India School of Computer Science and Engineering Galgotias University Uttar Pradesh Greater Noida203201 India School of Computer Science Engineering and Information Systems Vellore Institute of Technology Tamil Nadu Vellore632014 India Department of Communication and Computer Engineering Faculty of Engineering and IT Taiz University Taiz Yemen Katra India Department of Computer Science Engineering Jain University Bangalore India

Unstructured Numerical Image Dataset Separation (UNIDS) method employing an enhanced unsupervised clustering technique. The objective is to delineate an optimal number of distinct groups within the input grayscale (G-S) image content, utilizing the improved limited iteration agglomerative clustering (iLIAC) clustering technique for the separation and enhancement of picture elements. The UNIDS method is structured into two primary stages: partitioning and validation. In the partitioning stage, the UNIDS method identifies an appropriate number of discrete clusters within the grayscale image using the iLIAC technique, eliminating the need for predetermined procedures. Subsequently, the method evaluates the similarity and deviation among data elements within the same group in the resultant image dataset. Additionally, it assesses the proximity and inters severance among clusters in the outcome of the image dataset through the partitioning process. Empirical results indicate that the UNIDS system excels in the spontaneous identification of an optimal number of discrete clusters within the input G-S image. The system demonstrates superior thickness, reduced deviation among data elements within the same cluster, increased inter-separation, and diminished inter-closeness between cluster elements. Furthermore, empirical analysis establishes the superior performance of the UNIDS approach compared to existing clustering techniques. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

AutYOLO-ATT: an attention-based YOLOv8 algorithm for early autism diagnosis through facial expression recognition

引用

Neural Computing and Applications 2024年第27期36卷 17199-17219页

作者： Hosney, Reham Talaat, Fatma M. El-Gendy, Eman M. Saafan, Mahmoud M. Computers and Control Systems Engineering Department Faculty of Engineering Mansoura University Mansoura Egypt Faculty of Artificial Intelligence Kafrelsheikh University Kafrelsheikh33516 Egypt Faculty of Computer Science and Engineering New Mansoura University Gamasa35712 Egypt Nile Higher Institute for Engineering and Technology Mansoura Egypt

Autism Spectrum Disorder (ASD) is a developmental condition resulting from abnormalities in brain structure and function, which can manifest as communication and social interaction difficulties. Conventional methods for diagnosing ASD may not be effective in the early stages of the disorder. Hence, early diagnosis is crucial to improving the patient's overall health and well-being. One alternative and effective method for early autism diagnosis is facial expression recognition since autistic children typically exhibit distinct facial expressions that can aid in distinguishing them from other children. This paper provides a deep convolutional neural network (DCNN)-based real-time emotion recognition system for autistic kids. The proposed system is designed to identify six facial emotions, including surprise, delight, sadness, fear, joy, and natural, and to assist medical professionals and families in recognizing facial expressions in autistic children for early diagnosis and intervention. In this study, an attention-based YOLOv8 (AutYOLO-ATT) algorithm for facial expression recognition is proposed, which enhances the YOLOv8 model's performance by integrating an attention mechanism. The proposed method (AutYOLO-ATT) outperforms all other classifiers in all metrics, achieving a precision of 93.97%, recall of 97.5%, F1-score of 92.99%, and accuracy of 97.2%. These results highlight the potential of the proposed method for real-world applications, particularly in fields where high accuracy is essential. © The Author(s) 2024.

关键词： Diagnosis

来源：评论

学校读者我要写书评

暂无评论

Latency Aware Adaptive Ant Colony Algorithm for Service Placement for Healthcare Fog

引用

SN computer Science 2024年第8期5卷 1143页

作者： Goswami, Ankur Modi, Kirit Patel, Chirag Department of Computer Engineering Faculty of Engineering and Technology Sankalchand Patel Univeisity Gujarat Visnagar 384315 India Department of Computer Engineering Charuset University Gujarat Changa India

Fog computing offers a compelling paradigm for real-time healthcare data processing by minimizing latency and bringing computation closer to its source. However, efficient service placement remains a critical challenge for maximizing fog computing’s benefits in this domain. Existing service placement algorithms often struggle to adapt to dynamic fog environments and prioritize low latency for real-time healthcare applications. This research addresses this gap by proposing LA-AACO (Latency Aware Adaptive Ant Colony Optimization), a novel service placement algorithm specifically designed for healthcare applications in fog computing environments. LA-AACO incorporates an adaptive Latency Weight (β) parameter to balance exploration and exploitation during the search process. Additionally, it utilizes a latency-aware fitness function that directly prioritizes solutions with minimal overall latency for healthcare data processing. The LA-AACO is evaluated against established algorithms GWO and CSA with an ECG event monitoring application as the representative healthcare workload. The results demonstrate LA-AACO's superiority across all evaluated metrics, achieving significantly higher fog resource utilization ((93%), lower latency (0.19s), faster response (3.7s), lower energy consumption (1.5J) and faster runtime (39.4s) compared to existing algorithms. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Application placement Fog computing Healthcare applications

来源：评论

学校读者我要写书评

暂无评论

Hybrid Vision Transformer-Boosting Model for Skin Lesion Classification on SLICE-3D 12

Hybrid Vision Transformer-Boosting Model for Skin Lesion Cla...

引用

12th RSI International Conference on Robotics and Mechatronics, ICRoM 2024

作者： Teymouri, Sobhan Larki, Arash Abbasi Mohammadi, Melika Klidbary, Sajad Haghzad K.N. Toosi University of Technology Faculty of Electrical Engineering Tehran Iran University of Zanjan Faculty of Computer Engineering Zanjan Iran University of Zanjan Faculty of Engineering Department of Computer Engineering Zanjan Iran

ISBN: (纸本)9798331529734

Early detection of cutaneous malignancies remains a critical challenge in healthcare. This study proposes an approach to skin lesion classification using the SLICE-3D dataset, containing over 400,000 smartphone-quality dermatological images from more than 1,000 patients. Several models, such as the Dataefficient Image Transformer 3 (DeIT3) and Light Gradient Boosting Machine (LGBM), are employed in the proposed methodology. An ensemble approach is also utilized to capitalize on their collective capabilities and strengths. To address class imbalance, a multi-phase preprocessing protocol involving stratified sampling, Synthetic Minority Over-sampling Technique (SMOTE), and data augmentation is implemented. Quantitative analysis indicates that the combined model exhibits improved performance, achieving an accuracy of 89% and a recall of 90% for class 1. This outperforms the individual models: DeIT3 achieved an accuracy of 73 % with a recall of 91% for class 1, while LGBM achieved a higher accuracy of 96% but with a lower recall of 71% for class 1. These findings underscore the potential of ensemble methodologies in enhancing the accuracy and reliability of dermatological diagnostics for cutaneous malignancy detection. © 2024 IEEE.

关键词： Dermatology

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：