检索结果-内蒙古大学图书馆

Information Technology (ACIT)

作者： Anum Ilyas Surayya Obaid Narmeen Zakaria Bawany Department of Computer Science and Software Engineering Center for Computing Research Jinah University for Women Karachi Pakistan

In recent years, the use of CCTV footage for proactive crime prevention has surged, particularly in public places like airports, train stations, and malls. However, the efficacy of these surveillance systems becomes questionable for their substantial reliance on human resources which may lead to erroneous or delayed responses. This research takes a comprehensive approach for violence detection which was previously limited to binary classification, by categorizing violence into four distinct classes: abuse, arson, assault, and fight. It employs transfer learning combined with computer vision techniques for violence detection and classification in video footages. The study compares the performance of four pre-trained neural networks, namely DenseNet121, VGG16, MobileNet, and Xception. The dataset created for this research is compiled from three datasets available on Kaggle. The results reveal that the Xception model performed comparatively better achieving the highest AUC score of 98.39%, while the VGG16 model attained the lowest AUC score of 96%. In addition to the AUC score, precision, recall, and fl-score are employed as performance metrics. Transfer learning with convolutional neural networks (CNN) significantly reduced computational requirements and time. Automating the detection and categorization of violent behavior through the employed approach has the potential to reduce the risk of fatalities and injuries in public areas. It also improves the speed and accuracy of threat detection, enabling swift preventive actions by authorities.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Formal Verification Based Synthesis for Behavior Trees 9th

Formal Verification Based Synthesis for Behavior Trees

引用

9th International Symposium on Dependable software engineering: Theories, Tools and Applications, SETTA 2023

作者： Hong, Weijiang Chen, Zhenbang Li, Minglong Li, Yuhan Huang, Peishan Wang, Ji College of Computer National University of Defense Technology Changsha China Key Laboratory of Software Engineering for Complex Systems National University of Defense Technology Changsha China Institute for Quantum Information & State Key Laboratory of High Performance Computing National University of Defense Technology Changsha China

ISBN: (纸本)9789819986637

Behavior trees (BTs) have been extensively applied in the area of both computer games and robotics, as the control architectures. However, the construction of BTs is labor-expensive, time-consuming, and even impossible as the complexity of task increases. In this work, we propose a formal verification based synthesis method to automatically construct BTs whose behaviors satisfy the given Linear Temporal Logic (LTL) specifications. Our method first explores candidate BTs by a grammar-based Monte Carlo Tree Search (MCTS), then the explored BTs are transformed into Communicating Sequential Processes (CSP) models. After that, we invoke the verifier to check the models’ correctness w.r.t. specifications, and provide feedback based on the verification result for guiding the search process. The application of our method on several representative robotic missions indicates its promising. © 2024, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Behavior Trees CSP MCTS Synthesis

来源：评论

学校读者我要写书评

暂无评论

Investigation on Analog and Digital Modulations Recognition Using Machine Learning Algorithms

引用

World Journal of engineering and Technology 2024年第4期12卷 867-884页

作者： Jean Ndoumbe Ivan Basile Kabeina Gaelle Patricia Talotsing Soubiel-Noël Nkomo Biloo Laboratory of Computer Engineering Data Science and Artificial Intelligence National Higher Polytechnic School of Douala Douala Cameroon Mobile Computing and Networking Research Laboratory Department of Computer and Software Engineering Polytechnique Montreal Montreal Canada

In the field of radiocommunication, modulation type identification is one of the most important characteristics in signal processing. This study aims to implement a modulation recognition system on two approaches to machine learning techniques, the K-Nearest Neighbors (KNN) and Artificial Neural Networks (ANN). From a statistical and spectral analysis of signals, nine key differentiation features are extracted and used as input vectors for each trained model. The feature extraction is performed by using the Hilbert transform, the forward and inverse Fourier transforms. The experiments with the AMC Master dataset classify ten (10) types of analog and digital modulations. AM_DSB_FC, AM_DSB_SC, AM_USB, AM_LSB, FM, MPSK, 2PSK, MASK, 2ASK, MQAM are put forward in this article. For the simulation of the chosen model, signals are polluted by the Additive White Gaussian Noise (AWGN). The simulation results show that the best identification rate is the MLP neuronal method with 90.5% of accuracy after 10 dB signal-to-noise ratio value, with a shift of more than 15% from the k-nearest neighbors’ algorithm.

关键词： Automatic Recognition Artificial Neural Networks K-Nearest Neighbors Machine Learning Analog Modulations Digital Modulations

来源：评论

学校读者我要写书评

暂无评论

Buffer with Feature Tokens: Combating Confusion in Medical Image Continual Learning

Buffer with Feature Tokens: Combating Confusion in Medical I...

引用

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Zheming Fan Weiting Chen Jiahao Fan Ruizhi Yu Shanghai Key Laboratory of Trustworthy Computing MOE Research Center of Software/Hardware Co-Design Engineering East China Normal University

ISBN: (数字)9798350386226

ISBN: (纸本)9798350386233

Due to its complexity and privacy concerns, medical data is often difficult to collect fully at once. New data emerges with the discovery of new diseases and advances in medical technology, but privacy concerns limit its storage. When deep learning models require training on datasets of new diseases, they often suffer from catastrophic forgetting, where the models’ performance on previously learned data degrades significantly. Recently, pre-trained continual learning methods have been proposed. Unlike earlier continual learning methods, these new approaches do not require learning excessive amounts of new knowledge, furthermore, they can still achieve good performance even when the pre-trained feature extractor is completely frozen. For these approaches, the performance degradation is not primarily due to catastrophic forgetting, but rather confusion. To combat confusion, we propose a method using a buffer with feature tokens (BWFT) which is cheaper and safer than directly storing the original data. Additionally, we propose a feature merge method to further reduce storage and computation costs. In the class-incremental scenario, our method outperforms state-of-the-art methods on multiple medical datasets as well as natural image datasets. Code is available at https://***/1345475677/BWFT.

关键词： Continuing education Degradation Training Data privacy Privacy Costs Feature extraction Data models Biomedical imaging Diseases

来源：评论

学校读者我要写书评

暂无评论

Exploring Deep Learning and Grad-CAM for Speech-Based Detection of Mild Traumatic Brain Injury

Exploring Deep Learning and Grad-CAM for Speech-Based Detect...

引用

2024 IEEE International Conference on Big Data, BigData 2024

作者： Rojas, Fredy Madanian, Samaneh Templeton, John Michael Poellabauer, Christian Schneider, Sandra L. Department of Computer Science and Software Engineering Auckland New Zealand University of South Florida Department of Computer Science and Engineering TampaFL United States Florida International University School of Computing and Information Sciences MiamiFL United States Saint Mary's College Department of Communicative Sciences & Disorders Notre DameIN United States

ISBN: (纸本)9798350362480

Mild traumatic brain injury (mTBI) is challenging to diagnose due to its subtle and transient symptoms, making noninvasive diagnostic tools crucial for early detection. This study explores the use of a custom ResNet deep learning model combined with the Grad-CAM interpretability technique for mTBI detection via speech analysis. Speech data were transformed into Mel-spectrograms and fed into the model for binary classification between concussed and control individuals. The Grad-CAM method provided insights into which frequency regions of the Mel-spectrogram were most important for the model's predictions, with higher-frequency regions identified as significant for the model in detecting mTBI. Using Monte Carlo Cross-Validation (MCCV), we evaluated 50 different subject train-test split configurations to gain insights into the model's performance stability and variability. This analysis can assess the model's ability to learn consistent patterns within the dataset and suggest potential generalization tendencies. The variability observed in the performance metrics distribution underscores the importance of robust evaluation methods, particularly when working with small datasets. The combination of deep learning, robust evaluation technique and interpretability in this study contributes to the development of clinically viable, speech-based tools for mTBI detection, with potential applications in sports and healthcare settings. © 2024 IEEE.

关键词： Speech analysis

来源：评论

学校读者我要写书评

暂无评论

Identification of Molecular Biomarkers and Key Pathways among Idiopathic pulmonary fibrosis (IPF), Chronic Obstructive Pulmonary Disease (COPD) and Lung cancer 26

Identification of Molecular Biomarkers and Key Pathways amon...

引用

26th International Conference on Computer and Information Technology, ICCIT 2023

作者： Yasmin, Mst. Farjana Faruk Hosen, Md. Basar, Md. Abul Alam Shadhin, Khairul Trisha, Arifa Ferdoushi Shahin Uddin, Muhammad Mawlana Bhashani Science and Technology University Department of Information and Communication Technology Tangail1902 Bangladesh Department of Computing and Information System Dhaka Ashulia Bangladesh Department of Computer Science and Engineering Kanchon Purbachal American City1460 Bangladesh Department of Software Engineering Dhaka Ashulia Bangladesh

ISBN: (纸本)9798350359015

Lung cancer (LC), idiopathic pulmonary fibrosis (IPF) and chronic obstructive pulmonary disease (COPD) are the most fatal disorders in the globe, generating frequent human issues. Having IPF and COPD are the risk factors of the LC, but the molecular mechanisms that underlie among IPF, COPD, and LC are not yet elucidated. In this study, we looked for shared molecular indicators and pathways that might explain how individuals with IPF, COPD, and LC are related to one another. GSE24206, GSE76925 and GSE18842 microarray datasets are utilized for IPF, COPD and LC samples. The preprocessing of datasets has done using R language, and then concordant differentially expressed genes (DEGs) are discovered. After that the protein-protein interactions (PPIs) are built using the similar DEGs, and the hub genes are determined using topological analysis. ETS1, MSH2, SORD, RORA and NEDD9 are the PPI network's top 5 hub genes. The pathways of KEGG demonstrated that the concordant DEGs are related to the colorectal cancer and pathways in cancer. Future work for this project will focus on miRNA, TF, and gene ontology (GO) analyses, as well as module analysis networks. Finally, based on the concordant DEGs, a number of potential medications have been suggested. © 2023 IEEE.

关键词： COPD Differential expressed genes Drug compounds enrichment analysis Hub gene IPF Lung cancer Protein-protein network

来源：评论

学校读者我要写书评

暂无评论

Towards the Utilization of Parallel Programming to Speed Up RAM Forensics 10

Towards the Utilization of Parallel Programming to Speed Up ...

引用

10th International Conference on software Defined Systems, SDS 2023

作者： Al-Sharif, Ziad A. Alhamouri, Rahaf Jararweh, Yaser Lewis University Dept. of Engineering Computing and Mathematical Sciences Romeoville United States Jordan University of Sci. and Tech. Dept. of Software Engineering Irbid Jordan Jordan University of Sci. and Tech. Dept. of Computer Science Irbid Jordan

ISBN: (纸本)9798350382235

Memory forensics uses volatile digital artifacts as evidence about criminal activities. Analyzing captured memory dumps for volatile data requires time and effort. This paper studies the utilization of parallel programming to speed up RAM forensics. It presents a performance-based evaluation of parallel programming in the domain of memory forensics and compares sequential and parallel approaches to speed up the memory analysis process. First, it evaluates the sequential approach and uses it as a base case for further parallel approaches. Second, it evaluates two of the parallel approaches that can be performed on a typical user machine. Our experiments evaluate the use of two parallel programming paradigms: the in-process parallelization approach using OpenMP, and the inter-process parallelization approach using MPI. Our results compare the performance of the sequential approach, OpenMP thread-based approach, and MPI process-based approach. Experimentations compare the performance of three scenarios using six files of different sizes and various numbers of threads and/or processes. The results show that the use of MPI is slightly better than OpenMP approaches on the use of 2 and 4 processes/threads. However, when the number of processors/threads is increased to 8 and 16, OpenMP slightly outperforms the MPI approach. Additionally, the parallelization approach using OpenMP and MPI provides 3X to 5X speed up over the traditional sequential approach. Moreover, it is worth mentioning that this speed-up is achieved on traditional user machines without the use of HPC computers. © 2023 IEEE.

关键词： Parallel programming

来源：评论

学校读者我要写书评

暂无评论

A Novel Model-Driven Approach for Visual Impaired People Assistance OPTIC ALLY 3

A Novel Model-Driven Approach for Visual Impaired People Ass...

引用

3rd International Conference on Latest Trends in Electrical engineering and computing Technologies, INTELLECT 2022

作者： Rana, Laiba Rehman, Attique Ur Javaid, Sabeen Ali, Tahir Muhammad University of Sialkot Faculty of Computing and It Department of Software. Engineering Sialkot Pakistan Gulf University for Science and Technology Department of Computer Sciences Mubarak Al-Abdullah Kuwait

ISBN: (数字)9781665487061

ISBN: (纸本)9781665487061

People having visual impairment can't perform routine tasks on their own such that they are obliged to perform even the simplest task with assistance. Since their issue can't be resolved with visual glasses or lenses, the various mode of advancing technologies is playing an immense role in aiding them. Plenty of research work and development of gadgets have been proposed regarding the assistance of visually impaired people. From deep learning along with sensor-based initiative systems to various ranges of software have been developed but until now every system or software heeds on a particular feature. This article is based on purposing the idea of developing a cross-platform application named Optic Ally, intelligent assistance for visually impaired people that will provide all salient features in it regarding their assistance. For this idea of Optic Ally application, a novel model-driven framework is introduced that comprises a meta-model, tree editor, and graphical modeling tool that is Sirius based with drag and drop palette. The modeling and visualization of the features of this app are done using the Sirius-based graphical modeling tool. Moreover, a case study has been demonstrated for the validity of the proposed framework. The outcome of the case study proves that the proposed framework is competent in modeling and visualizing features effectively. The initial developed stage of the application is also shown in this article. Optic Ally application features include object detection, object finder, location, voice assistance, guardian help, link account and profile setup for the visually impaired user whereas for the guardian user the features include location, profile setup and guardian help © 2022 IEEE.

关键词： Object detection

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

Facial Expression Recognition Based on TripletLoss and Attention Mechanism

Facial Expression Recognition Based on TripletLoss and Atten...

引用

International Conference on Automation and computing (ICAC)

作者： Rongqiang Gou Qiuyan Gai Zhijie Xu Yuanping Xu Chaolong Zhang Jin Jin Jia He Yajing Shi School of Software Engineering Chengdu University of Information Technology Chengdu China School of Computing and Engineering University of Huddersfield Huddersfield UK

The paper proposes a facial expression recognition method called Triplet-Loss Attention Network (TAN), which aims to address the problem of large intra-class and inter-class distances in facial expression recognition. The method uses a self-attention mechanism to calculate the weight of triplet expression samples, which helps in determining the influence of key regions on expression recognition. The method uses images with attention weights above a certain threshold as training samples to form new hard triplets. The distance between high-weight samples in the three sets of samples is calculated using the Mahalanobis distance formula, and the difference in distance between high-weight groups of triplets is calculated using the Maharaja Loss function. The Triplet Loss is mainly used as the loss function in TAN, and the model is jointly optimized using Triplet Loss, Mahalanobis Loss, and Cross Entropy Loss functions to improve the performance of the model in facial expression recognition. Experimental results show that TAN performs well in alleviating the intra-class and inter-class distance problem and has good robustness and generalization performance. On the RAF-DB and FERPlus datasets, TAN achieves recognition accuracies of 88.40% and 88.73%, respectively, which is 1.37% and 0.18% higher than the previous state-of-the-art methods.

关键词：

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：