检索结果-内蒙古大学图书馆

Multidomain Object Detection Framework Using Feature Domain Knowledge Distillation

IEEE Transactions on Cybernetics 2024年第8期54卷 4643-4651页

作者： Jaw, Da-Wei Huang, Shih-Chia Lu, Zhi-Hui Fung, Benjamin C. M. Kuo, Sy-Yen National Taiwan University Department of Electrical Engineering Taipei106 Taiwan National Taipei University of Technology Department of Electronic Engineering Taipei106 Taiwan Fudan University School of Computer Science The Engineering Research Center of Cyber Security Auditing and Monitoring Ministry of Education Shanghai200433 China McGill University School of Information Studies MontrealQCH3A 1X1 Canada

Object detection techniques have been widely studied, utilized in various works, and have exhibited robust performance on images with sufficient luminance. However, these approaches typically struggle to extract valuable features from low-luminance images, which often exhibit blurriness and dim appearence, leading to detection failures. To overcome this issue, we introduce an innovative unsupervised feature domain knowledge distillation (KD) framework. The proposed framework enhances the generalization capability of neural networks across both low- and high-luminance domains without incurring additional computational costs during testing. This improvement is made possible through the integration of generative adversarial networks and our proposed unsupervised KD process. Furthermore, we introduce a region-based multiscale discriminator designed to discern feature domain discrepancies at the object level rather than from the global context. This bolsters the joint learning process of object detection and feature domain distillation tasks. Both qualitative and quantitative assessments shown that the proposed method, empowered by the region-based multiscale discriminator and the unsupervised feature domain distillation process, can effectively extract beneficial features from low-luminance images, outperforming other state-of-the-art approaches in both low- and sufficient-luminance domains. © 2013 IEEE.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Privacy protection and data security in intelligent recommendation systems

引用

Neural Computing and Applications 2025年 1-18页

作者： Zheng, Donghua Alkawaz, Mohammed Hazim Johar, Md Gapar Md School of Information Technology and Engineering Guangzhou College of Commerce Guangdong Guangzhou511363 China School of Graduate Studies Management and Science University Selangor Shah Alam40100 Malaysia Department of Computer Science College of Education for Pure Science University of Mosul Nineveh Mosul41002 Iraq Software Engineering and Digital Innovation Centre Management and Science University Selangor Shah Alam40100 Malaysia

Data security and privacy have emerged as major areas of study for recommendation systems. This is crucial for enhancing system security given the quick advancement of Internet technology and its increased use. However, there are still issues with privacy protection and data security in current intelligent recommendation systems, such as a high probability of privacy leakage, low data security, and poor recommendation performance. To better promote the wider application of intelligent recommendation systems, this article introduces federated learning and differential privacy technology to conduct in-depth research on privacy protection and data security in intelligent recommendation systems. In the article, a federated learning framework is used to construct an intelligent recommendation system model, and the constructed model is trained. Furthermore, a differential privacy mechanism is introduced, which can protect the privacy of samples by analyzing the sensitivity of L1 and L2, gradient descent, gradient pruning, and other methods. Next, the secure multi-party computation (SMC) protocol can be used to achieve model parameter sharing and protect the privacy of participants;subsequently, there is the implementation of federated learning. Finally, this article also evaluates the practical application effects of federated learning and differential privacy technology in privacy protection and data security in intelligent recommendation systems. The study’s findings indicate that the approach suggested in this article has an average data security and privacy leakage risk of 5.35% and 96.64%;federated learning is 16.55% and 85.08%, respectively;differential privacy is 9.65% and 88.47%, respectively;deep learning is 21.08% and 77.92%, respectively;and homomorphic encryption is 10.69% and 82.43%, respectively. The combination of federal learning and differential privacy can effectively protect the privacy and data security of the intelligent recommendation system, reduce

关键词： Sensitive data

来源：评论

学校读者我要写书评

暂无评论

CNN-Based Deep Learning Model for Early Identification and Categorization of Melanoma Skin Cancer Using Medical Imaging

引用

SN computer Science 2024年第7期5卷 911页

作者： Pandimurugan, V. Ahmad, Sultan Prabu, A.V. Rahmani, Mohammad Khalid Imam Abdeljaber, Hikmat A. M. Eswaran, Malathi Nazeer, Jabeen Computer Science and Engineering Acharya Institute of Technology Bangalore India Department of Computer Science College of Computer Engineering and Sciences Prince Sattam Bin Abdulaziz University Alkharj 11942 Saudi Arabia University Center for Research and Development (UCRD) Department of Computer Science and Engineering Chandigarh University Gharuan Punjab Mohali 140413 India Department of Electronics and Communication Engineering Koneru Lakshmaiah Education Foundation Vaddeswaram AP Guntur 522502 India College of Computing and Informatics Saudi Electronic University Riyadh Saudi Arabia Department of Computer Science Faculty of Information Technology Applied Science Private University Amman Jordan MEU Research Unit Middle East University Amman 11831 Jordan Department of Computer Technology Kongu Engineering College Perundurai Tamil Nadu Erode India

In human life, skin cancer is a curse. If not appropriately diagnosed, it spreads around all body parts in the earlier stage. The melanoma skin cancer death rate is 75% all over the world. There is an urgent need for a cure to be in place. Swarm and evolutionary computation for healthcare deals with real-world healthcare applications. One of the best ways to do this is to find the genes that need to be turned off such that only cancer cells can die. The current therapies used to cure cancer focus on turning off cancer cells but, at times, affect healthy tissues. Using the power of Artificial Intelligence, one can easily mark out the cancer-affected area and provide treatment accordingly. Segmentation techniques can be used to mark cancer lesions, and advanced deep learning algorithms in computer vision can help classify which class of cancer lesions. Automatic detection of melanoma by using a dermoscopic sample is tricky to find the stage and percentage of lesions affected by using deep convolutional neural networks with the help of machine vision tools. To predict skin lesions, we developed a model with three layers, each having output channels of 16, 32, and 64, respectively. For this research, different samples were collected from the international skin imaging collaboration database (ISIC2019, ISIC2020, ISIC2021). The study computed the most important parameters for classifying and identifying the lesion with accuracy, precision, recall, and specificity values. The proposed DCNN model classifier attained an accuracy of 88.82%, 93.45%, and 95.15% with the respective datasets of ISIC2019, ISIC2020, and ISIC2021, indicating high performance compared with the other state-of-the-art networks. The proposed model is a better framework for automating early detection and classification of melanoma skin cancer to protect many lives and avoid mishaps. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Classification Convolution neural network Deep learning Medical imaging Melanoma Skin cancer Swarm and evolutionary computation

来源：评论

学校读者我要写书评

暂无评论

Adaptive Varying Contention Window MAC Protocol Based on Underwater Acoustic Propagation Delay

引用

IEEE Sensors Journal 2023年第13期23卷 15019-15031页

作者： Guo, Yongan Obaidat, Mohammad S. Guo, Jing Wu, Tin-Yu Tey, Fu Jie Engineering Research Center of Health Service System Based on Ubiquitous Wireless Networks Ministry of Education College of Telecommunications and Information Engineering Nanjing210003 China The King Abdullah II School of Information Technology University of Jordan Amman11942 Jordan The School of Computer and Communication Engineering University of Science and Technology Beijing Beijing100083 China Amity University UP Noida201301 India Nanjing University of Posts and Telecommunications College of Telecommunications and Information Engineering Nanjing210003 China National Pingtung University of Science and Technology Department of Management Information Systems Pingtung912301 Taiwan National Taiwan University of Science and Technology Department of Electrical Engineering Taipei106335 Taiwan

Underwater acoustic wave propagation time delay and fluctuations are much greater than radio waves. However, many of the protocols used for underwater acoustic networking and communication are designed according to the principles of terrestrial wireless communication networks, failing to give full consideration to utilizing the underwater acoustic propagation and its time delay variation. For this reason, this article proposes an adaptive varying contention window media access control (AVCW-MAC) protocol based on propagation delay. The protocol uses AVCW random backoff mechanism to differently set up the CW of each node based on the different propagation delays from each underwater sensor node to the cluster head node to improve the channel utilization. In addition, to avoid excessive difference in the CW of each node, the backoff factors K1 and K2 are introduced in the AVCW random backoff mechanism to improve the fairness of nodes' accessing the channel. Simulation results show that for the same amount of sensor nodes, the AVCW-MAC protocol has significantly improved network throughput and decreased average end-to-end latency when compared to legacy CW-MAC protocols. © 2001-2012 IEEE.

关键词： Underwater acoustics

来源：评论

学校读者我要写书评

暂无评论

Real-Time Web Application to Classify Diabetic Foot Ulcer 14

Real-Time Web Application to Classify Diabetic Foot Ulcer

引用

14th International Conference on Computing Communication and Networking Technologies, ICCCNT 2023

作者： Nandi, Souhrid Anurag, Apoorva Mayya, Veena Jeganathan, Jayakumar Manipal Academy of Higher Education Manipal Institute of Technology Department of Computer Science & Engineering Technology Karnataka Manipal India Manipal Academy of Higher Education Manipal Institute of Technology Department of Information & Communication Technology Karnataka Manipal India Manipal Academy of Higher Education Kasturba Medical College Dept. of Medicine Karnataka Mangalore India

ISBN: (纸本)9798350335095

Diabetic foot ulcers are one of the major complications of type 2 diabetes and can result in gangrene and amputation if left untreated. Early detection and treatment are essential for reducing the risk of amputation. Imaging, such as digital photography or imaging scans, can be a valuable tool in detecting and monitoring diabetic foot ulcers, documenting their size and location, identifying underlying issues, and evaluating treatment effectiveness. This paper utilises a two-dimensional image as its input and applies an ensemble of neural networks to analyse the images. The system is capable of performing image classification into four distinct states, infection, ischaemia, both, and none. © 2023 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Security Enhancement Using Blockchain for Internet of Things 11th

Security Enhancement Using Blockchain for Internet of Things

引用

11th International Conference on Intelligent Computing and Applications, ICRTC 2023

作者： Dhamodaran, S. Talukdar, Veera Ramesh, J.V.N. Sardana, Kamal Talukdar, Suryansh Bhaskar Sathappan, S. Department of Computer Science and Engineering Sathyabama Institute of Science and Technology Chennai India RNB Global University Rajasthan Bikaner India Department of Computer Science and Engineering Koneru Lakshmaiah Education Foundation Andhra Pradesh Guntur Vaddeswaram India Department of Electronics and Communication Engineering Technological Institute of Textile and Sciences Haryana Bhiwani India School of Computer Science and Engineering VIT Bhopal Madhya Pradesh Bhopal India Department of Information Technology Chaitanya Bharathi Institute of Technology Andhra Pradesh Hyderabad Gandipet India

ISBN: (纸本)9789819717231

This research concentrates on the security concerns raised by the IoT. There have been many research conducted on the topic of security;however, the issue with these studies is that they are not particularly effective. In addition, the results of these examinations did not provide an acceptable level of protection. To save both time and storage space, the suggested work employs a lossless compression approach to compress data. After that, information is encrypted and stored on the blockchain. To store encrypted and compressed data over a blockchain, a hash key is first used to initialize the nonce and timestamp. Then, the receiver takes the data it needs from the blockchain and decompresses and decrypts it. Outputs of simulations present hybrid blockchain mechanism that is incorporating compression and encryption, is significantly increasing the level of security. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Cryptography

来源：评论

学校读者我要写书评

暂无评论

Improving Tomato Disease Diagnosis Through ResNet50: An Innovative Approach 4

Improving Tomato Disease Diagnosis Through ResNet50: An Inno...

引用

4th International Conference on Sustainable Expert Systems, ICSES 2024

作者： Dhamotharan, R. Raj, J. Relin Francis Krishnan, R. Santhana Kavitha, S. Soundiraraj, N. Kumar, V. Vinoth Kalasalingam Academy of Research and Education Department of Computer Science and Information Technology Krishnankoil India Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Electronics and Communication Engineering Chennai India Scad College of Engineering and Technology Department of Electronics and Communication Engineering Cheranmahadevi India Velammal College of Engineering and Technology Department of Computer Science and Engineering Madurai India Psna College of Engineering and Technology Department of Electrical and Electronics Engineering Dindigul India Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology Department of Electronics and Communication Engineering Chennai India

ISBN: (纸本)9798331540364

Tomato crops are vulnerable to numerous diseases that can severely affect both yield and quality. Early and precise identification of these diseases is indispensable for implementing active management and intervention measures. This research introduces an advanced approach to tomato disease detection by employing the ResNet50 architecture, pretrained on the ImageNet dataset. Through transfer learning, the ResNet50 model leverages its pre-learned features to classify tomato plant images affected by diseases such as Tomato Mosaic Virus (ToMV), Bacterial Spot, Blossom End Rot, Early Blight, and Late Blight. The ResNet50 model is fine-tuned with custom classification layers tailored to the specific diseases, allowing it to achieve high accuracy in disease detection. Preliminary results indicate that the ResNet50 model significantly outperforms traditional machine learning approaches, demonstrating its efficacy in accurately diagnosing tomato diseases. This study provides an effective solution for improving disease detection in tomatoes and highlights the transformative role of deep learning in advancing agricultural practices, fostering sustainability, and enhancing productivity. © 2024 IEEE.

关键词： Fruits

来源：评论

学校读者我要写书评

暂无评论

Seastar: A Cache-Efficient and Load-Balanced Key-Value Store on Disaggregated Memory

Seastar: A Cache-Efficient and Load-Balanced Key-Value Store...

引用

2024 IEEE International Conference on Cluster Computing, CLUSTER 2024

作者： Du, Jingwen Wang, Fang Feng, Dan Zeng, Dexin Yi, Sheng Engineering Research Center of Data Storage Systems and Technology Ministry of Education of China Wuhan National Laboratory for Optoelectronics Key Laboratory of Information Storage System China School of Computer Science and Technology Huazhong University of Science and Technology China Shenzhen Huazhong University of Science and Technology Research Institute China

ISBN: (纸本)9798350358711

In modern datacenters, memory disaggregation un-packs monolithic servers to build independent network-connected compute and memory pools, greatly improving resource uti-lization. Existing memory-disaggregated key-value stores adopt ownership sharing or ownership partitioning for request scheduling and they all show poor performance and scalability. The former generates multiple cache misses and multiple network round trips due to poor cache locality, while the latter fails to fully use resources on all Compute Nodes (CNs) due to load imbalance. In this paper, we propose Seastar, a fast and scalable key-value store that achieves high cache locality as well as load balancing. To enable high concurrency and ensure linearizability, Seastar exploits a self-verification version chain to provide out-of-place atomic updates and ensure consistent reads. To improve cache locality, Seastar leverages a pair-based ownership assigning scheme to assign the processing permission of requests corre-sponding to each key to a CN pair through two independent hash functions. To ensure load balancing among CNs, Seastar proposes a decentralized power-of-two-choices routing scheme to route the requests to the less-loaded CNs in CN pairs for handling. Experimental results demonstrate that Seastar improves the throughput by up to 6.7x and significantly reduces the latency compared with state-of-the-art memory-disaggregated key-value stores. © 2024 IEEE.

关键词： Scalability

来源：评论

学校读者我要写书评

暂无评论

Integrity of Code and IoT Validation of Resource Utilization in Micro Control Unit

Integrity of Code and IoT Validation of Resource Utilization...

引用

2023 International Conference on New Frontiers in Communication, Automation, Management and Security, ICCAMS 2023

作者： Kushwaha, Ritesh Kumar Ravikumar, Parimi Lvd Mohammed, Muzammil H Sasthi Kumar, C. Chandigarh University Department of Electronics & Communication Engineering Punjab Mohali India Shri Vishnu Engineering College for Women Deparmnent of Information Technology Bhimavaram India Taif University Department of Information Technology Taif Saudi Arabia Karpagam Academy of Higher Education Department of Computer Science Coimbatore India

ISBN: (纸本)9798350317060

Cybersecurity for the Internet of Things is gaining more and more attention as the number of IoT devices has increased and the number of attack stargeting these devices continues to grow exponentially. However, the significant focus areas for security measures have been guaranteeing services' constant availability and protecting users' private data. But code integrity is not receiving the attention it deserves because these devices lack the horsepower to provide more robust safeguards. This issue necessitates more study into adapting standard procedures for code integrity to IoT gadgets. This paper describes a virtual machine-based technique for ensuring the correct execution of code on the Internet of Things. The feasibility of a virtual machine tailored to safeguard software during firmware upgrades is demonstrated here. The suggested method's low resource requirements on popular microcontrollers have been verified experimentally. The acquired findings prove that this secured technique for coding may work in practice. © 2023 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Enforcing Social Distancing with YOLO Algorithm Utilizing Object-to-Object Distance 24

Enforcing Social Distancing with YOLO Algorithm Utilizing Ob...

引用

24th International Arab Conference on information technology, ACIT 2023

作者： Irsan, Muhamad Hassan, Rosilah Abdali, Taj-Aldeen Naser Ishak, Mohamad Khairi School of Computing Telkom University Bandung Indonesia College of Engineering and Information Technology Ajman University Department of Information Technology United Arab Emirates University of Misan Faculty of Basic Education Misan62001 Iraq Ajman University Department of Electrical and Computer Engineering Ajman United Arab Emirates

ISBN: (纸本)9798350384307

Instances of COVID-19 transmission occur daily due to individuals failing to maintain distance or engaging in physical contact with others who may be contaminated with the virus. To mitigate this issue, this study has developed a system to detect human subjects practicing social distancing. The system utilizes a Raspberry Pi 4 Model B 8GB device in combination with a Logitech HD Webcam C270 camera. To detect human subjects, the Convolutional Neural Network is employed, utilizing the You Only Look Once (YOLO) method. In the testing phase of the tool, the system successfully identifies human subjects and assesses their proximity to others. It also detects instances of social distancing violations. The system achieved an average mean Average Precision (mAP) of 0.9792, a Precision of 0.9482, a Recall of 0.9819, and an f1 score of 0.9648. © 2023 IEEE.

关键词： COVID Image Raspberry Pi Social distancing YOLO

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：