检索结果-内蒙古大学图书馆

Semantic Segmentation Based Real-Time Traffic Monitoring via Res-UNet Classifier and Kalman Filter

SN computer science 2025年第1期6卷 1-17页

作者： Qureshi, Asifa Mehmood Algarni, Asaad Aljuaid, Hanan Alatiyyah, Mohammed Hamad Alnowaiser, Khaled Jalal, Ahmad Department of Creative Technologies Air University E-9 Islamabad 44000 Pakistan Department of Computer Sciences Faculty of Computing and Information Technology Northern Border University Rafha 91911 Saudi Arabia Department of Computer Sciences College of Computer and Information Sciences Princess Nourah Bint Abdulrahman University P.O. Box 84428 Riyadh 11671 Saudi Arabia Department of Computer Science College of Sciences and Humanities in Aflaj Prince Sattam Bin Abdulaziz University Al-Kharj Saudi Arabia Department of Computer Engineering College of Computer Engineering and Sciences Prince Sattam Bin Abdulaziz University Al-Kharj 11942 Saudi Arabia Department of Computer Science Air University E-9 Islamabad 44000 Pakistan

In computer vision, accurate vehicle recognition and tracking is a challenging research topic. Manual surveillance systems are cumbersome, labor-intensive, and inefficient in today's congested traffic environment. Automated traffic monitoring systems are needed in today's world because they provide potential solutions to these issues. However, there are still some challenges in automating these systems which need to be addressed. One of the challenges is the extraction of foreground from a complex environment that only contains vehicles along with spatial and temporal information. In this paper, a semantic segmentation-based foreground extraction model is proposed to overcome these challenges. After segmentation, a blob detection technique is applied to detect each vehicle. To overcome the issue of vehicle-to-vehicle occlusion we extracted Harris corner points and computed the separation line based on the vehicle geometry coordinates. Furthermore, we introduce an ID assignment mechanism based on VGG-16 extracted features. Based on the detections, the number of cars is also counted for each frame. For tracking, we first applied Farneback optical flow algorithm to detect motion in vehicles. Only those vehicles were passed to the tracking algorithm in which motion was detected. To estimate each vehicle's location in succeeding image frames Kalman filter was implemented. A vehicle reidentification and ID recovery module are also proposed to enhance the tracking outcome. Vehicle trajectories based on vehicle location midpoint were also estimated. Extensive experiments are used to validate the proposed system and achieved a precision score of 0.965 and 0.931 over Roundabout and AU-AIR datasets for vehicle detection. However, for vehicle tracking the precision is 0.948 over the Roundabout dataset and 0.839 for the AU-AIR dataset. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Blob detection Detection and tracking Kalman filter Occlusion handling Res-Unet Semantic segmentation

来源：评论

学校读者我要写书评

暂无评论

Identification of Ayurvedic Plants Using Deep Learning

Identification of Ayurvedic Plants Using Deep Learning

引用

2024 IEEE International Conference on Computing, Semiconductor, Mechatronics, Intelligent Systems and Communications, COSMIC 2024

作者： Kambli, Prashanth Satrasala, Vikas Shivanna, K. Shreyash Sharma, Shaan Neralkar, Yash Nitin M S Ramaiah Institute of Technology Information Science & Engineering Bangalore India Computer Science & Engineering Sahyadri College of Engineering and Management Adyar Mangalore India

ISBN: (纸本)9798331517892

Deep learning algorithms have become pivotal in the accurate identification and classification of Ayurvedic plants, leveraging advanced techniques to distinguish between various species based on leaf images. Among these algorithms, the VGG16 model, a convolutional neural network (CNN), is particularly notable for its robust performance in image recognition tasks. The Ayurvedic plant detection database serves as a comprehensive resource for training and validating these models, facilitating high accuracy in plant detection. The integration of deep learning models presents significant potential for botany and health care applications. By offering reliable and precise identification of Ayurvedic plants, the proposed system supports botanists and healthcare professionals in making informed decisions. This leads to more accurate identification and utilization of traditional medicinal knowledge. These advancements not only improve the efficiency and reliability of plant detection systems but also build confidence in AI-driven solutions within the field of traditional medicine. The potential to revolutionize plant classification and enhance the understanding of medicinal plants underscores the importance of continued research and innovation in this domain. The model proposed ensures faster and easier way to recognize the medicinal plants and give its description to the reader using deep learning methods with the accuracy of 96% identification systems. © 2024 IEEE.

关键词： Medicinal chemistry

来源：评论

学校读者我要写书评

暂无评论

Development of Control Programming and Data Analytics on The Controlled Automatic Sorting Simulation System 7

Development of Control Programming and Data Analytics on The...

引用

7th International Conference on information technology, InCIT 2023

作者： Bundasak, Supaporn Chomchan, Iya Kasetsart University Sriracha Campus Faculty of Science at Sriracha Computer Science and Information Technology Department Chonburi Thailand

ISBN: (纸本)9798350358698

The Faculty of science, Sriracha Campus is in the Eastern Economic Corridor (EEC) to recognize the importance of the development of Industry4.0, Therefore it has been developed a control and data analysis program based on a controlled automatic sorting simulation system to disseminate the technology of PLC and data analysis to develop into a smart factory. By developing personnel to have knowledge and developing the capabilities of machines for Industry4.0. Automated machines for smart factories, such as production systems, can be adjusted according to the situation by responding in real time. Moreover, by creating intelligent machines, trackers and forecasters can analyze production data by receiving information from machines to prevent errors from occurring. Being able to develop a machine status monitoring program can help extend the working life of machines. Therefore, developing a training set to program and study data analysis in factory machinery. It is important to develop technological knowledge and understanding in developing smart factory systems such as automatic control. Machine maintenance through data analysis from various sensors using production data to analyze production rates or detecting work controls through sensors that have complex functions from various research developments through studies on this simulation set in the future. © 2023 IEEE.

关键词： Data Analytics

来源：评论

学校读者我要写书评

暂无评论

A System Review on Fraudulent Website Detection Using Machine Learning Technique

引用

SN computer science 2023年第6期4卷 702页

作者： Saraswathi, P. Anchitaalagammai, J.V. Kavitha, R. Department of IT Velammal College of Engineering and Technology Madurai India Department of Computer Science Engineering Velammal College of Engineering and Technology Madurai India

At present, scams and malicious websites are one of the most widespread and dangerous problems on the website. It brings enormous economic suffering and irretrievable losses to companies and individuals. This approach can strengthen the Internet’s legitimacy and impose sanctions on criminals who engage in prohibited or malicious activities. However, governments still need a derivation to classify websites as dangerous or non-dangerous. However, several malicious and counterfeit goods are published on fraudulent websites to cheat consumers and make high and unfair profits. Due to the proliferation of such fraudulent websites, it is difficult to detect and identify them through manual inspection. Phishing attacks include various attacks, including spoofing malicious-based, DNS-based, data theft, email/spam, web-based delivery, and telephone-based phishing. We propose an integrated machine learning (ML) framework for fraudulent website detection to solve this problem. Artificial neural networks (ANN), support vector machine (SVM), random forests (RF), and K-nearest neighbor (K-NN) are algorithms to detect phishing websites accurately. Some URLs can be used to classify them as appropriate or phishing. Data from publicly available phishing websites can be collected from the UCIrvine ML repository for training and testing. Then, the results can be predicted using the features of the dataset. We conduct an in-depth literature review and propose methods for detecting phishing websites using ML methods. © 2023, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： ANN DNS KNN ML Phishing RF SVM UCIrvine URL

来源：评论

学校读者我要写书评

暂无评论

Automated Detection of Infection in Diabetic Foot Ulcer Using Pre-trained Fast Convolutional Neural Network with U++net

引用

SN computer science 2024年第6期5卷 705页

作者： Murthy, S.V.N. Bhargavi, Kovvuri N. Isaac, Samson N.Ganesh, E. Department of Computer Science and Engineering SJC Institute of Technology Information Technology Aditya College of Engineering and Technology Biomedical Engineering Karunya Institute of Technology and Sciences Department of ECE St. Peter’s Institute of Higher Education and Research

A frequent consequence of diabetes and a significant contributor to morbidity and mortality is diabetic foot ulcer (DFU).Early detection and appropriate management of DFUs are essential to prevent complications such as infections and lower extremity amputations. In recent years, medical imaging and machine learning have emerged as promising tools for the automated detection and analysis of DFUs. We gathered a sizable foot imaging dataset, including DFU from multiple patients. This paper proposes a novel preprocessing technique based on the Shades of Gray color constancy algorithm to cope with noise and lighting variations in diabetic foot ulcer (DFU) images captured from different devices. The algorithm aims to enhance image quality, improve illumination normalization, and mitigate the impact of noise, thus providing more reliable and accurate DFU analysis and detection. Using the Diabetic Foot Infection Network with the Adam Optimizer (DFINET-AO), features were retrieved after the dataset had been preprocessed and divided. In order to comprehend the normal and pathological spectrum of diabetes, image data and numerical/text data are separated independently. Foot images of patients with aberrant diabetes coverage are separated from each other and classified using Pre-trained Fast Convolutional Neural Network (PFCNN), which has been trained on the U++network. Classification techniques, like foot ulcer analysis, forecast a etiology. This study's primary goal was to establish a novel method for evaluating the likelihood that diabetes individuals may acquire foot ulcers by imaging analysis of existing foot ulcers. The data was preprocessed and segmented after the researchers gathered a collection of foot photographs and medical information from historical records of diabetes patients. The amount of normal and pathological diabetes was then determined from numerical and textual data by extracting characteristics from the segmented data using DFINET-AO. To detect foot ulc

关键词： Diabetic foot infection network with the adam optimizer Diabetic foot ulcer Pre-trained fast convolutional neural network

来源：评论

学校读者我要写书评

暂无评论

An Intelligent Dipsomania Face Detection Model using Spatial and Spectral Graph Neural Network 5

An Intelligent Dipsomania Face Detection Model using Spatial...

引用

5th International Conference on Smart Electronics and Communication, ICOSEC 2024

作者： Madhusundar, N. Surendran, R. Johnpaul, S. Appathurai, K. Saveetha Institute of Medical and Technical Sciences Department of Computer Science and Engineering Chennai India Saveetha Institute of Medical and Technical Sciences Saveetha School of Engineering Department of Computer Science and Engineering Chennai602105 India Vel Tech Multi Tech Dr. Rangarajan Dr. Sakunthala Engineering College Department of Information Technology Chennai India Kg College of Arts and Science Department of Computer Technology Coimbatore India

ISBN: (纸本)9798331504403

Dipsomania, alcohol addiction that is out of control poses serious problems for public health. Conventional techniques for identifying drunk people depend on behavioral indicators, which are easily manipulated. This article suggests a novel method by employing spectral and spatial graph neural networks (SS-GNN) to create an intelligent face identification model for dipsomania. Facial traits are stored as nodes and edges in a graph structure, and the model uses the special powers of graph neural networks to process and analyses these features. The SS-GNN model that has been suggested is trained on a wide range of facial image datasets that include both drunken and sober states. Advanced methods of data augmentation and domain adaptation are utilized to enhance the model's resilience and applicability to various demographics and environmental circumstances. By combining spectral and spatial information into a graph-based framework, dipsomania-related facial traits can be better understood. This innovative method provides a noninvasive, dependable means of early diagnosis and intervention in cases of alcohol addiction and shows potential for real-time applications in public safety, healthcare, and workplace monitoring. © 2024 IEEE.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Belief Combination of Classifiers for Incomplete Data

引用

IEEE/CAA Journal of Automatica Sinica 2022年第4期9卷 652-667页

作者： Zuowei Zhang Songtao Ye Yiru Zhang Weiping Ding Hao Wang IEEE the Research Center for Optical Fiber Sensing Zhejiang Laboratory the School of Automation Northwestern Polytechnical University the School of Cyber Engineering Xidian University the Department of Computer Science CY Cergy Paris Université the School of Information Science and Technology Nantong University the School with of Cyber Engineering Xidian University the Department of Computer Science Norwegian University of Science and Technology

Data with missing values,or incomplete information,brings some challenges to the development of classification,as the incompleteness may significantly affect the performance of *** this paper,we handle missing values in both training and test sets with uncertainty and imprecision reasoning by proposing a new belief combination of classifier(BCC)method based on the evidence *** proposed BCC method aims to improve the classification performance of incomplete data by characterizing the uncertainty and imprecision brought by *** BCC,different attributes are regarded as independent sources,and the collection of each attribute is considered as a ***,multiple classifiers are trained with each subset independently and allow each observed attribute to provide a sub-classification result for the query ***,these sub-classification results with different weights(discounting factors)are used to provide supplementary information to jointly determine the final classes of query *** weights consist of two aspects:global and *** global weight calculated by an optimization function is employed to represent the reliability of each classifier,and the local weight obtained by mining attribute distribution characteristics is used to quantify the importance of observed attributes to the pattern *** comparative experiments including seven methods on twelve datasets are executed,demonstrating the out-performance of BCC over all baseline methods in terms of accuracy,precision,recall,F1 measure,with pertinent computational costs.

关键词： Classifier fusion classification evidence theory incomplete data missing values

来源：评论

学校读者我要写书评

暂无评论

Adaptive Quantum Learning Frameworks for Real-Time IIoT Attack Identification 1

引用

7th International Conference on Intelligent Computing and Optimization, ICO 2023

作者： Joshi, Poonam Goyal, S.B. Rajawat, Anand Singh Solanki, Ram Kumar Chen, Chuqiao Department of Computer Engineering Thakur College of Engineering and Technology Mumbai India Faculty of Information Technology City University Petaling Jaya46100 Malaysia School of Computer Science & Engineering Sandip University Nashik422213 India

ISBN: (数字)9783031733185

ISBN: (纸本)9783031733178

Securing its networks from cyber-attacks is of utmost importance as the Industrial Internet of Things (IIoT) becomes a lynchpin of contemporary industrial ecosystems. With the increasing complexity and sophistication of cyber threats, traditional machine learning approaches, while successful, frequently struggle with real-time identification due to computational restrictions. In this study, we present an adaptive quantum learning framework for real-time IIoT attack identification that takes advantage of quantum computing’s inherent parallelism and speed. We introduce a novel quantum technique that can quickly and accurately identify both new and existing attack vectors as they emerge in dynamic threat landscapes. The experimental results show improved accuracy rates and a drastic decrease in identification delay compared to classical methods. According to our research, quantum-enhanced learning frameworks have great potential to strengthen IIoT security in the face of more sophisticated cyber-attacks. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Cyber attacks

来源：评论

学校读者我要写书评

暂无评论

Resolving Attached USBs: Analysis of Windows 11 Artifacts 2

Resolving Attached USBs: Analysis of Windows 11 Artifacts

引用

2nd International Conference for Advancement in technology, ICONAT 2023

作者： Budhrani, Akash Singh, Upasna Singh, Bhupendra Defence Institute of Advanced Technology School of Computer Engineering and Mathematical Science Pune India Indian Institute of Information Technology Computer Science and Engineering Department Pune India

ISBN: (纸本)9781665475174

The Windows Operating System is known for its convenience which tends to breed more and more user information in form of Artifacts. Artifacts are important repository of potential evidence while conducting any computer investigation. This paper thus provides an analysis of artifacts that will bring out the inside of USB devices attached or mounted on a particular system. Moreover, this paper will parse the data out into usable information and helps to interpret the findings of forensic examiner. In this work, Image file of the suspected system has been exported from the system for further analysis to bring out relevant information. © 2023 IEEE.

关键词： Windows operating system

来源：评论

学校读者我要写书评

暂无评论

Distortion minimization for multimedia transmission in NOMA HAP-UAV integrated aerial access networks

引用

Chinese Journal of Aeronautics 2022年第9期35卷 81-94页

作者： Guobin ZHANG Zhu HAN Hang XIN Xin WEI Boya DI School of Electrical Engineering and Intelligentization Dongguan University of TechnologyDongguan 523808China Department of Electrical and Computer Engineering and Department of Computer Science University of HoustonTX77004USA School of Computer Science and Technology Dongguan University of TechnologyDongguan 523808China Department of Computing Imperial CollegeLondon SW72AzUK

Aerial access networks have been envisioned as a promising 6G solution to enhance the ground communication systems in both coverage and capacity. To better utilize the spectrum and fully explore different channel characteristics, this paper constructs an integrated network comprising the High Altitude Platform(HAP) and Unmanned Air Vehicles(UAVs) with the NonOrthogonal Multiple Access(NOMA) technology. In order to improve the transmission quality of images and videos, a power management scheme is proposed to minimize the distortion of the transmissions from the HAP and UAVs to the terminals. The power control is formulated as a non-convex problem constrained by the maximal transmit power and the minimal terminal rate requirements. The variable substitution and the first-order Tailor’s expansion is used to transform it into a sequence of convex problems, which are subsequently solved through the gradient projection method. Simulation demonstrates the signal distortion and error rate improvement achieved by the proposed algorithm.

关键词： Aerial Access Networks High Altitude Platform Non-Orthogonal Multiple Access Sixth-Generation Unmanned Air Vehicles

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：