检索结果-内蒙古大学图书馆

Automated Area Based Explosive Trace Detection System

Automated Area Based Explosive Trace Detection System

2024 International Conference on science, engineering and Business for Driving Sustainable Development Goals, SEB4SDG 2024

作者： Ohemu, Monday F. Azeta, Ambrose A. Adeyanju, Ibrahim A. Maidoki, Nuradeen ACETEL-National Open University of Nigria Department of Artiftial Intelligent Nigeria Namibia University of Science and Technology Department of Software Engineering Namibia Federal University Department of Computer Engineering Oye-Ekiti Nigeria Natview Technology Abuja Nigeria

ISBN: (纸本)9798350358155

Explosive based attacks on people and sensitive places in the form of terrorism has become a global challenge that is making organizations such as airports, train station, security agencies and the government to do anything possible to secure the lives of people and some essential infrastructures. Several approaches have been adopted to prevent its occurrence that ranges from the use of animals, analytical means and sensor based approach to manual screaming by human personnel. With the development in the field of Artificial Intelligent, efforts are being made to deploy Sensor base system that leverage of machine learning technology to make the system highly accurate and sensitive in detection of explosive trace within an environment. This study presents the architecture of area base explosive trace detection utilizing deep learning model. The experiment shows the concept of area based explosive trace detection from trace properties of Carbon, Nitrogen, Oxygen and Hydrogen (CNOH). The deep leaning model recorded an accuracy of 98% on the explosive trace data. The system uses this information to detect the presence of explosion within an area. © 2024 IEEE.

关键词： Airport security

来源：评论

学校读者我要写书评

暂无评论

Weapon Detection Identification and Classification for DCNN and YOLO-V5 Techniques 5

Weapon Detection Identification and Classification for DCNN ...

引用

5th International Conference on IoT Based Control Networks and Intelligent Systems, ICICNIS 2024

作者： Ashok Kumar, J. Dhivya, G. Selvaraj, P. Karthikeyan, V. Ganesh Ramesh, S.P. Lakshmi, S. Jeya Saravanakumar, R. Karpagam Academy of Higher Education Department of Computer Applications Tamil Nadu Coimbatore India School of Computer Science and Engineering Department of Computer Science and Engineering Galgotias University Uttar Pradesh Greater Noida India School of Computing SASTRA Deemed to Be University Thanjavur Tamil Nadu Thirumalaisamudram India Iconix Software Solution Tamil Nadu Tirunelveli India

ISBN: (纸本)9798331518097

Over 850,000 people die every year as a direct result of gun violence, yet civilians hold more than 85% of the world's weapons. Detecting weapons via manual surveillance has not been successful. It is critical to automate a system that detects weapons. A large number of convolution neural network CNN-based automated weapon identification systems have shown promising results. These methods are crucial for weapon identification but expensive. These devices produce a large amount of false negatives since monitoring videos are not high enough quality or visible enough to identify weapons. Finding the sweet spot between fast detection and a few false positives and negatives in weapon detection is the main goal of this research. In addition to defining an Area of Interest, the suggested structure uses YOLO. The models use previously processed images with backgrounds that have been blurred using a median filter. Metrics such as F1 scores, accuracy, recall, False Positive, and False Negative will be used to assess the suggested architecture. This investigation shows that YOLO-v5 can quickly identify and has a high recall rate. The deep convolution neural network DCNN required 0.17 ms for each frame, but this one only required 0.010 ms, which would indicate that security and weapon detection have good promise. Starting with a labeled collection of weapon and non-weapon photos is the initial stage in DCNN and YOLOv5 weapon detection. Use YOLOv5's ability to recognize multiple things within frames to train a real-time object localization and identification model. DCNNs categorize weapons and extract complex data effectively. DCNN classifies objects, whereas YOLOv5 detects, bounds, and tracks them. Optimize both models' hyperparameters and apply data augmentation for better performance. Install the integrated system on cutting-edge equipment for real-time weapon identification and classification. © 2024 IEEE.

关键词： Median filters

来源：评论

学校读者我要写书评

暂无评论

Meta Pattern Concern Score: A Novel Evaluation Measure with Human Values for Multi-classifiers

Meta Pattern Concern Score: A Novel Evaluation Measure with ...

引用

2023 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2023

作者： Wang, Yanyun Du, Dehui Liu, Yuanhao The University of Hong Kong Department of Computer Science Hong Kong Software Engineering Institute East China Normal University Shanghai China

ISBN: (纸本)9798350337020

While advanced classifiers have been increasingly used in real-world safety-critical applications, how to properly evaluate the black-box models given specific human values remains a concern in the community. Such human values include punishing error cases of different severity in varying degrees and making compromises in general performance to reduce specific dangerous cases. In this paper, we propose a novel evaluation measure named Meta Pattern Concern Score based on the abstract representation of probabilistic prediction and the adjustable threshold for the concession in prediction confidence, to introduce the human values into multi-classifiers. Technically, we learn from the advantages and disadvantages of two kinds of common metrics, namely the confusion matrix-based evaluation measures and the loss values, so that our measure is effective as them even under general tasks, and the cross entropy loss becomes a special case of our measure in the limit. Besides, our measure can also be used to refine the model training by dynamically adjusting the learning rate. The experiments on four kinds of models and six datasets confirm the effectiveness and efficiency of our measure. And a case study shows it can not only find the ideal model reducing 0.53% of dangerous cases by only sacrificing 0.04% of training accuracy, but also refine the learning rate to train a new model averagely outperforming the original one with a 1.62% lower value of itself and 0.36% fewer number of dangerous cases. © 2023 IEEE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Recognition of Fire Situation Using Graph Convolutional Network Model 15th

Recognition of Fire Situation Using Graph Convolutional Netw...

引用

15th International Conference on computer science and its Applications, CSA 2023 and 17th KIPS International Conference on Ubiquitous Information Technologies and Applications, CUTE 2023

作者： Kim, Si Jin Park, Ji Su Kang, Jungho Shon, Jin Gon Advanced Convergence Technology Laboratory Korea National Open University Seoul Korea Republic of Department of Computer Science and Engineering Jeonju University Jeonju-Si Korea Republic of Department of Software Engineering Baewha Woman University Seoul Korea Republic of Department of Computer Science Korea National Open University Seoul Korea Republic of

ISBN: (纸本)9789819724468

Fires cause a lot of casualties and economic losses. In order to prevent fire accidents in advance, it is necessary to find out the cause of the fire. Existing fire alarm systems detected fires with temperature, smoke, and flame sensors, but could not distinguish the cause of the fire. In this paper, to solve this problem, the recognition of fire situation was tested using the GCN model. The fire dataset used in the experiment is a fire dataset composed of causes such as boxes, clothing, and electricity. The accuracy of 93.44% was confirmed as a result of the learning experiment with the GCN model, a graph-based deep learning model. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Premixed flames

来源：评论

学校读者我要写书评

暂无评论

A Study on the Application of Protein Language Models in the Analysis of Membrane Proteins 19th

A Study on the Application of Protein Language Models in...

引用

19th International Symposium on Distributed Computing and Artificial Intelligence, DCAI 2022

作者： Ghazikhani, Hamed Butler, Gregory Department of Computer Science and Software Engineering Concordia University Montreal Canada Centre for Structural and Functional Genomics Concordia University Montreal Canada

ISBN: (纸本)9783031232091

Membrane proteins make up around 30% of all proteins in a cell. These proteins are difficult to evaluate due to their hydrophobic surface and dependence on their original in vivo environment. There is a tremendous demand for computational approaches for predicting membrane proteins’ function and structure. Protein sequences, the language of life, can benefit from using natural language algorithms. In this study, we propose to use the BERT language model to represent membrane proteins. Our preliminary results, which utilized a Logistic Regression classifier, reveal that BERT can comprehend the context of membrane proteins and enable us to investigate better and distinguish these proteins. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Proteins

来源：评论

学校读者我要写书评

暂无评论

Color Image Steganalysis Algorithm Based on Quaternion Convolutional Neural Network 30th

Color Image Steganalysis Algorithm Based on Quaternion Convo...

引用

30th International Conference on Computational and Experimental engineering and sciences, ICCES 2024

作者： Xu, Meng Wang, Jinwei State Key Laboratory of Mathematical Engineering and Advanced Computing Henan450001 China Department of Computer and Software Nanjing University of Information Science and Technology Nanjing210044 China

ISBN: (纸本)9783031687747

Most of the images on the Internet are color images, and steganalysis of color images is a very critical issue in the field of steganalysis. The current proposed color image steganalysis features mainly rely on manual design, and the steganalysis features do not fully consider the internal connections between the three channels of color images. Most steganalyzers based on deep learning are also designed for grayscale images. In recent years, the continuous introduction of advanced steganography techniques for color images has also brought more severe challenges to the steganography analysis of color images. Quaternions are a tool that can effectively represent color images, and utilizing quaternion transformations can fully exploit the correlation between color image channels. Combining quaternion theory with convolutional neural networks, this paper proposed a color image steganalysis algorithm based on quaternion convolutional neural networks. Firstly, the image is represented by quaternion, and then input into the quaternion convolutional neural network. The convolution kernel in the network is composed of quaternions, and then perform convolution and pooling. The experimental results have demonstrated the significant effectiveness of the proposed network. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Managing Information Updating with Edge Computing: A Distributed and Learning Approach 48

Managing Information Updating with Edge Computing: A Distrib...

引用

48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

作者： He, Junyi Zhang, Di Liu, Shumeng Zhou, Yuezhi Zhang, Yaoxue Beijing Jiaotong University School of Software Engineering China The Chinese University of Hong Kong School of Science and Engineering Shenzhen China Tsinghua University Department of Computer Science and Technology China

ISBN: (纸本)9781728163277

The rapid proliferation of some real-time applications (e.g., video surveillance) has driven enormous interest in maximizing information freshness, quantified by the age of information (AoI). For some computation-intensive updates such as images or videos, the real-time update processing requires intensive resources, which edge servers can provide in mobile edge computing (MEC). In this paper, we investigate information updating scheduling with multiple users in MEC. Due to the centralized algorithms' limitations in distributed systems where users are self-interested, we investigate an efficient distributed scheduling algorithm. We model the information updating scheduling as an uncooperative game and propose a distributed algorithm to compute the unique Nash equilibrium. Considering the unavailability of some global network information, we propose a learning algorithm where each user learns how to make decisions based on observable information in a distributed manner. Extensive evaluation results show the efficiency of the proposed algorithms. © 2023 IEEE.

关键词： Age of information distributed updating scheduling mobile edge computing

来源：评论

学校读者我要写书评

暂无评论

Abnormal Behavior Detection Using Deep-Learning-Based Video Data Structuring

引用

Intelligent Automation & Soft Computing 2023年第8期37卷 2371-2386页

作者： Min-Jeong Kim Byeong-Uk Jeon Hyun Yoo Kyungyong Chung Department of Computer Science Kyonggi UniversitySuwon-si16227Gyeonggi-doKorea Contents Convergence Software Research Institute Kyonggi UniversitySuwon-si16227Gyeonggi-doKorea Division of AI Computer Science and Engineering Kyonggi UniversitySuwon-si16227Gyeonggi-doKorea

With the increasing number of digital devices generating a vast amount of video data,the recognition of abnormal image patterns has become more ***,it is necessary to develop a method that achieves this task using object and behavior information within video *** methods for detecting abnormal behaviors only focus on simple motions,therefore they cannot determine the overall behavior occurring throughout a *** this study,an abnormal behavior detection method that uses deep learning(DL)-based video-data structuring is *** and motions are first extracted from continuous images by combining existing DL-based image analysis *** weight of the continuous data pattern is then analyzed through data structuring to classify the overall *** performance of the proposed method was evaluated using varying parameter settings,such as the size of the action clip and interval between action *** model achieved an accuracy of 0.9817,indicating excellent ***,we conclude that the proposed data structuring method is useful in detecting and classifying abnormal behaviors.

关键词： Deep learning object detection abnormal behavior recognition classification data structuring

来源：评论

学校读者我要写书评

暂无评论

Gland Segmentation in H&E Histopathological Images using U-Net with Attention Module 12

Gland Segmentation in H&E Histopathological Images using U-N...

引用

12th International Conference on Image Processing Theory, Tools and Applications, IPTA 2023

作者： Yildirim, Zeynep Samet, Refik Hancer, Emrah Nemati, Nooshin Mali, Mohamed Traore Ankara University Department of Computer Engineering Ankara06830 Turkey Mehmet Akif Ersoy University Department of Software Engineering Burdur15030 Turkey Victoria University of Wellington School of Engineering and Computer Science Wellington6140 New Zealand

ISBN: (纸本)9798350325416

Gland segmentation refers to the process of identifying and delineating glandular regions within histopathology images. However, gland segmentation in histopathology images is a challenging task due to several factors such as shape variations and overlapping structures. Despite numerous research efforts dedicated to addressing this challenge, it remains an unresolved problem. To address this issue, we introduce a deep gland segmentation method by designing an enhanced U-Net variant combined with an Attention module. By incorporating the Attention module into the U-Net architecture, the model becomes more adept at capturing contextual information and fine details within the histopathology images. To verify the effectiveness of the introduced method, we compare it with well-known semantic segmentation architectures on the GLAS and GRAG datasets. According to the results, our approach demonstrates a high success rate in gland segmentation, effectively competing with existing state-of-the-art methods. © 2023 IEEE.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Personalized Health Assistant with Reinforcement Learning 1

Personalized Health Assistant with Reinforcement Learning

引用

1st IEEE International Conference on Artificial Intelligence for Medicine, Health and Care, AIMHC 2024

作者： Jin, Jennifer Kim, Mira Kim, Soo Dong Computer Science and Engineering California State University San BernardinoCA92407 United States Department of Computer Science California State University FullertonCA92831 United States School of Software Soongsil University Seoul Korea Republic of

ISBN: (纸本)9798350371987

Integration of medical IoT devices with machine learning presents significant advantages for personalized healthcare. This research is to develop a software solution which leverages smart scale devices to measure health indexes, recommend personalized health-promoting activities. We designed the system with Q-learning elements and tailored the Q-learning policy to keep learning the personal effectiveness of health-promoting activities through users’ episodes. The system was implemented in Python, and the experiments demonstrated a substantial average enhancement of 25% across health indexes and 53% in optimizing the performance of the recommender policy. ©2024 IEEE.

关键词： Reinforcement learning

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：