检索结果-内蒙古大学图书馆

2nd International conference on Machine Learning and Autonomous systems, ICMLAS 2025

作者： Yang, xiaoqin Alias, Syazwina Wang, Hui FoEBEIT SEGi University Kota Damansara Selangor Malaysia College of Computer and Communication Engineering Nanjing Tech University Pujiang Institute Jiangsu Nanjing210000 China

ISBN: (纸本)9798331505745

This study proposes a vehicle image anomaly detection framework based on improved K-means algorithm and high-resolution camera, which provides a new idea for road safety construction from the image processing level. The main innovation of the study contains 4 aspects: 1. A new vehicle target extraction algorithm based on deep learning is proposed. The algorithm uses the YOLOv5s network structure and combines the SimAM attention mechanism to optimize the model while improving the detection accuracy of vehicle targets. 2. A visual-based traffic flow clustering analysis is performed. This analysis achieved effective clustering of traffic flow through deep HOG feature extraction and parallel K-means clustering algorithm. 3. An optimized license plate detection and recognition algorithm based on YOLOv5 is proposed. The algorithm improves the accuracy of license plate recognition by adding the key point detectors. 4. The traditional K-means algorithm is optimized. The optimization idea is mainly to combine quantum computing and space-time theory. This optimization improves the convergence speed and clustering effect of the algorithm. The results of comparative experiments show that the improved K-means algorithm converges faster in traffic flow clustering analysis, and its license plate recognition accuracy is significantly better than traditional algorithms (such as k-means and FCM). © 2025 IEEE.

关键词： K-means clustering

来源：评论

学校读者我要写书评

暂无评论

Adaptive enhancement algorithm of high grayscale X-ray films image based on statistical decision

引用

Jisuanji Jicheng Zhizao xitong/Computer Integrated Manufacturing systems, CIMS 2025年第3期31卷 888-902页

作者： Luo, Cheng Wang, Peng Li, Liangliang Sha, Baolin Lyu, Zhigang Li, xiaoyan Yang, Bo School of Electronics and Information Engineering Xi'an Technological University Xi'an710021 China School of Mechatronic Engineering Xi'an Technological University Xi'an710021 China The 41st Institute of the Fourth Academy CASC Xi'an710025 China Shaanxi Xiyu Nondestructive Testing Co. Ltd. Xianyang712199 China

Aiming at the problem that high-gray-scale X-ray negative images cannot be displayed directly on low-level monitors. an adaptive enhancement algorithm for high-grayscale X-ray negative images based on statistical decision-making was pro-posed. In view of the problem that 12bit high grayscale images could not be displayed, a RAW-Optical mapping algorithm was designed to quantize 12bit images to 16bit lossless display, For the problem that the gray value after mapping was too concen trated, the image was whitened and the contrast was too low, the upper and lower limits of the cropping variable were deter-mined according to the image quantization characteristics and the ARAW-Optical algorithm was designed to dynamically stretch the gray scale of the image to improve the image contrast. Taking advantage of the prominent feature of weld seam structure information of industrial X-ray films an automatic exposure discrimination model based on statistical decision-making was designed by introducing the average grayscale difference and average deviation, and an adaptive grayscale correction algo-rithm AGRAW-Optical was designed according to the exposure degree to achieve image brightness correction and enhance im age detail information. The experimental results showed that the proposed method could effectively realize the clear display of high gray-scale images, improve the visual perception of weld information, and enhance the interpretation and identification of industrial X-ray film welding slogans, welding dates and defect information. Especially in the low-illumination X-ray negative image processing, compared with the existing advanced algorithms, various indicators had been greatly improved. the informa tion entropy was increased by 0.58%, the effective efficiency of edge information was increased by 12. 75 times, the contrast was increased by 0,24%, and the image clarity index was improved nearly 10 times. © 2025 CIMS. All rights reserved.

关键词： X ray films

来源：评论

学校读者我要写书评

暂无评论

A Machine Learning Framework for Skin Cancer Detection using Transfer Learning

A Machine Learning Framework for Skin Cancer Detection using...

引用

International image processing, Applications and systems conference (IPAS)

作者： Mary Linda Panakal Augustine Sotirios Spanogianopoulos Mhd Saeed Sharif Department of Computer Science and Digital Technologies School of Architecture Computing and Engineering University of East London London UK School of Electrical and Mechanical Engineering University of Portsmouth Portsmouth UK

ISBN: (数字)9798331506520

ISBN: (纸本)9798331506537

This paper uses deep learning algorithms including InceptionV2, InceptionV3, DenseNet, MobileNet, and VGG19 to improve skin cancer detection. This research aims to improve skin cancer diagnosis. This work aims to determine the most effective evolutionary metrics-based technique to recognizing skin cancer, which is comparable to other diseases. Ultimately, our paper aims to create a realistic skin cancer detection system that uses the best deep learning algorithm. This discovery might improve medical diagnostics, leading to earlier diagnosis and improved healthcare outcomes.

关键词： Deep learning Machine learning algorithms image processing Transfer learning Medical diagnosis Reliability Skin cancer image classification

来源：评论

学校读者我要写书评

暂无评论

image-based System for Blood Group Classification

Image-based System for Blood Group Classification

引用

International conference on ent Computing and Control systems (ICICCS)

作者： K. Ramakrishna Reddy G. Indrani N. Pavan Kumar K. Vamshi Krishna Department of Computer Science and Engineering (AI & ML) Vignan Institute of Technology and Science Nalgonda Telangana

ISBN: (数字)9798331512088

ISBN: (纸本)9798331512095

It provides a classic method for blood image based system for blood group classification techniques. The suggested approach reduces reliance on traditional processes by automating the analysis of blood sample images through the use of sophisticated image processing algorithms. This strategy seeks to offer a dependable and expandable solution that is especially useful for application in distant or resource-constrained environments.

关键词： image segmentation Histograms Control systems Classification algorithms Blood

来源：评论

学校读者我要写书评

暂无评论

Soccer Player Tracking Using UAV imagery: A Comparative Study of Yolo and Traditional image processing algorithms

Soccer Player Tracking Using UAV Imagery: A Comparative Stud...

引用

International conference on Unmanned Aircraft systems (ICUAS)

作者： Felipe dos Anjos Rezende Thayron Miranda Hudson Pedro Augusto Fialho Silva Werikson Frederiko de Oliveira Alves André Luis Carvalho Mendes Alexandre Santos Brandão Nucleus of Specialization in Robotics Department of Electrical Engineering Universidade Federal de Viçosa Viçosa MG Brazil

ISBN: (数字)9798331513283

ISBN: (纸本)9798331513290

Player tracking is a useful tool for tactical analysis and performance evaluation in soccer, providing valuable insights into player movements and team dynamics. This project investigates the feasibility of tracking players using UAVcaptured imagery, employing both YOLO and traditional image processing algorithms (TIPA). Initial validation focuses on robot soccer players due to their predictable and controllable movements. The comparative analysis considers processing time, computational cost, adaptability to environmental changes, sensitivity to lighting variations, ability to handle dynamic conditions, tracking accuracy, and real-time performance. Results indicate that, under equivalent hardware and preparation time conditions, YOLO achieves performance comparable to traditional techniques. Nonetheless, the selection of the most suitable approach should be guided by task-specific demands, available computational resources, and the time allocated for system development and deployment.

关键词： YOLO Performance evaluation Sensitivity Tracking image processing Lighting Robot sensing systems Prediction algorithms Real-time systems Sports

来源：评论

学校读者我要写书评

暂无评论

Research on Video Content Analysis and Recommendation System Based on image processing Technology in New Media Environment

Research on Video Content Analysis and Recommendation System...

引用

Digital Analysis and processing, Intelligent Computation (DAPIC), International conference on

作者： Bei Zhang Anhui Professional College of Art Hefei Anhui China

ISBN: (数字)9798331536169

ISBN: (纸本)9798331536176

In the new media environment, the research of video content analysis and recommendation system is particularly important. This paper discusses how to use image processing technology to deeply analyze and accurately recommend film and television content. With the diversification of presentation forms and communication channels of film and television content, the problems of information overload and uneven content quality have become increasingly prominent. This study aims to solve these problems and improve the user experience by constructing a video content analysis and recommendation system based on image processing technology. The system includes four core modules: data acquisition, image processing, content analysis and recommendation algorithm. By integrating image processing, machine learning and recommendation algorithm, the whole process from data acquisition to personalized recommendation is implemented efficiently. The experimental results demonstrate that the hybrid recommendation strategy outperforms single algorithms in various evaluation metrics, including accuracy, recall, F1 score, and Mean Reciprocal Rank (MRR), significantly enhancing the precision of recommendations and user satisfaction. This study not only provides a scientific basis and decision support for content creation, distribution, and marketing in the film and television industry but also lays a solid foundation for building more intelligent film and television content recommendation systems.

关键词： Industries Content management TV Machine learning algorithms Accuracy image processing Data acquisition Media User experience Recommender systems

来源：评论

学校读者我要写书评

暂无评论

12th International conference on Recent Trends in Computing, ICRTC 2024

12th International Conference on Recent Trends in Computing,...

引用

12th International conference on Recent Trends in Computing, ICRTC 2024

ISBN: (纸本)9789819788354

The proceedings contain 34 papers. The special focus in this conference is on Recent Trends in Computing. The topics include: Deep Learning-Based Prediction of Human Ageing Related Proteins;the Insight of Clustering algorithms for Emerging Networking Paradigms;Revolutionizing Cardiovascular Health: Exploring the Potential of AI-Driven Approaches in Diagnosing Arrhythmia Disorders;humor Detection in Telugu Social Media Text Using Cost-Sensitive Learning and Indian Language Embeddings;empirical Comparison of Sleep Disorder Prediction Using Machine Learning Techniques;securing Smart Grids: Decentralized Anomaly Detection Using Federated Learning and Recurrent Neural Networks;machine Learning-Powered Intrusion Detection for Network Security;secure Chat System: Harnessing the Power of Hybrid Encryption for Enhanced Security and Confidentiality;deep Neural Networks for Early Diabetes Mellitus Prediction;advancements in image Deblurring and Performance Metrics Using Deep Learning Technique;emotion-Levitating Music Recommendation System Through Real-Time Facial Recognition;smartVoice TrashMate: The Intelligent Voice-Controlled Trash Bin;an Innovative image Enhancement Algorithm for Color Filter Array images: A Novel Approach;Enhancing Secure and Efficient Communication in Free-Space Optical Networks Using Spectral Amplitude Coding - OCDMA Technology;enhancing Citrus Fruit Classification and Quality Assessment Through Advanced Machine Learning Techniques;connecting Health for a Better Tomorrow Through Internet of Medical Things;image processing in Retail Marketing: Innovations for Customer Engagement, Sales, and Operational Efficiency;Leaf Checker: AI-Based Plant Diagnostic Tool Using Alexnet and Mobilenet;ioT-Supported Indoor Navigation System Using Machine Learning Technique;a Customizable Virtual Assistant Enhanced with Vision and Speech Capabilities;An Unequal Cluster-Based Routing Mechanism Using Lyrebird Optimization Algorithm to Achieve Network Longevity in WSNs.

关键词：

来源：评论

学校读者我要写书评

暂无评论

12th International conference on Recent Trends in Computing, ICRTC 2024

12th International Conference on Recent Trends in Computing,...

引用

12th International conference on Recent Trends in Computing, ICRTC 2024

ISBN: (纸本)9789819789450

关键词：

来源：评论

学校读者我要写书评

暂无评论

Perceptual Learning Technology Based on Computer Vision and Machine Learning algorithms 14

Perceptual Learning Technology Based on Computer Vision and ...

引用

14th IEEE International conference on Communication systems and Network Technologies, CSNT 2025

作者： Wang, Tianjian School of Computer Science and Technology Wuhan Qingchuan University Hubei Wuhan430204 China

ISBN: (纸本)9798331531935

To address the issue of low accuracy and poor robustness of perceptual learning in complex scenarios, a new method integrating computer vision and machine learning is adopted, that is, by applying deep neural networks, transfer learning and self-supervised learning, combined with multimodal data fusion strategy, to improve target recognition efficiency and learning ability. First, based on the preprocessing algorithm of improved Gaussian filtering and image segmentation, the quality of image features is improved. Data enhancement methods such as random rotation, flipping, and blurring are used to expand data distribution and improve model adaptability. Secondly, a convolutional neural network (CNN) is utilized in combination with an attention mechanism to extract multi-scale target features, and transfer learning is applied to transfer common features from pre-trained models to reduce dependence on large-scale labeled data. Finally, a contrastive learning framework is constructed to mine the correlation of unlabeled data. Transformer is used to realize the fusion of multimodal data of images and texts, and the model performance is optimized through multi-task learning. The mAP (Mean Average Precision) of the traditional method in dynamic occlusion scenarios is 60.5%, which is relatively weak. This may be because the traditional method cannot fully extract the effective features of the target under the occlusion of complex moving targets. The mAP of the method in this paper is 72.8%, which is higher than that of the traditional method. The perceptual learning method adopted in this paper effectively improves the accuracy and robustness in complex scenarios, and provides reliable technical support for intelligent applications. In the future, it can be combined with edge computing to further optimize the real-time processing capability and promote its application in the fields of unmanned driving and intelligent security. © 2025 IEEE.

关键词： image segmentation

来源：评论

学校读者我要写书评

暂无评论

Quantum Variational Singular Value Decomposition: A new hybrid approach towards quantum image compression

Quantum Variational Singular Value Decomposition: A new hybr...

引用

International Communication systems and Networks and Workshops, COMSNETS

作者： Britant Abhishek Tiwari CDAC Noida Uttar Pradesh India

ISBN: (数字)9798331531195

ISBN: (纸本)9798331531201

This research paper explores the application of singular value decomposition (SVD) in quantum image processing (QIP), specifically focusing on the computation of eigenvalues using variational quantum algorithms. SVD is a powerful mathematical tool in image processing, used for tasks such as image compression, noise reduction, and feature extraction. In this study, we propose quantum variational singular value decomposition (QVSVD) based on the variational quantum deflation (VQD) algorithm to determine the eigenvalues that contributes in calculating the singular values of the image matrix and eventually doing the SVD. This facilitates the extraction of eigenvalues with exponential speedup on real hardware compared to classical methods. We detail the implementation of this quantum algorithm within the framework of QIP, highlighting its advantages in terms of computational efficiency and accuracy on a Fault-Tolerant Quantum Computing (FTQC). Furthermore, we present a comparative analysis of the quantum and the classical method of SVD, demonstrating the accuracy of the image data. The simulation results validate the theoretical advantages of using quantum algorithms for SVD in image processing. This work not only underscores the potential of quantum computing in enhancing image processing techniques but also sets the stage for future research in the field, exploring more complex image processing tasks and other quantum algorithms. Our findings suggest that quantum image processing can offer unprecedented capabilities, paving the way for advancements in various applications such as medical imaging, remote sensing, and multimedia processing.

关键词： Training Quantum computing Quantum algorithm image coding Accuracy Feature extraction Eigenvalues and eigenfunctions Matrix decomposition Time complexity Singular value decomposition

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：