检索结果-内蒙古大学图书馆

11th European Workshop on Visual Information processing, EUVIP 2023

作者： Atabay, Habibollah Agh Hassanpour, Hamid Shahrood University of Technology Image Processing & Data Mining Lab Shahrood Iran

ISBN: (纸本)9798350342185

Anomaly detection in surveillance videos, particularly related to human behavior, is crucial for various applications. In electronic exams (e-exams), cheating can be detected using surveillance videos, but previous research mainly focused on predefined patterns, with less attention given to unsupervised methods. This study proposes a semi-supervised skeleton-based approach for abnormal behavior detection in e-exam proctoring videos. The proposed method segments the skeleton-based feature vectors of consecutive frames based on their similarity. The similarity is calculated based on the Euclidean distance between the mean feature vector and the standard deviation of segments. Similar segments of the training set, which is anomaly-free, are then combined to form distinct frame segments used to recognize the normal samples. Then, in the testing phase, a frame is recognized anomaly if it is not similar to any training segment, otherwise, it is considered normal. In addition, to get a better ranking of anomalous frames, we examine assigning a soft anomaly score considering the segment size and the maximum distance between the comparative features. In label assignment, the algorithm correctly categorizes the testing frame based on their similarity to the training segments. In comparison to the state-of-the-art reconstruction-based anomaly detection algorithms, the proposed method outperforms using the area under the ROC curve (AUC) metric. In addition to correctly detecting similar frames and segmenting videos, the advantage of the presented method is to determine the data labels without needing to process all inputs, making it possible for use in online applications. © 2023 IEEE.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Abnormal Behavior Detection in Electronic-Exam Videos Using BeatGAN 8

Abnormal Behavior Detection in Electronic-Exam Videos Using ...

引用

8th International Iranian Conference on Signal processing and Intelligent Systems, ICSPIS 2022

作者： Atabay, Habibollah Agh Hassanpour, Hamid Shahrood University of Technology Image Processing & Data Mining Lab Shahrood Iran

ISBN: (纸本)9781665476232

Abnormal Behavior Detection is a core ability to recognize cheating in electronic exams (e-exams), especially in the scenes where a fraudulent candidate hides unauthorized resources from the view of the proctor. This work utilizes the BeatGAN algorithm to detect behavioral abnormalities in the skeleton-based motion information of candidates during remote e-exams. We examine the algorithm on the previously published dataset of skeleton features extracted from e-exam videos. The results show that BeatGAN outperforms previously investigated algorithms in sequence anomaly detection using the raw joint-coordinate, implicit bone-length-and-angle and explicit bone-length feature types. But, in event anomaly detection, BeatGAN does not improve the state-of-the-art results on the dataset. © 2022 IEEE.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Feature Extraction for Polyp Detection in Wireless Capsule Endoscopy Video Frames

引用

Journal of Healthcare Engineering 2023年第1期2023卷

作者： Delagah, Bardia Hassanpour, Hamid Image Processing and Data Mining Lab Shahrood University of Technology Shahrud Iran

Wireless capsule endoscopy (WCE) is a technology for filming the gastrointestinal (GI) tract to find abnormalities such as tumors, polyps, and bleeding. This paper proposes a new method based on hand-crafted features to detect polyps in WCE frames. A polyp has a convex surface containing pixel values with a specified Gaussian distribution. If a polyp exists in the WCE image, edges will be seen at the border of the occupied area. Since WCE images often suffer from low illumination, a histogram equalization (HE) technique can be used to enhance the image. In this paper, we initially find probable polyp edges via thresholding. Then, we use the edges to find the region of interest (ROI). Then, the mean, standard deviation (STD), and division of mean by STD from the ROI are computed as features to discriminate between polyp and nonpolyp using a support vector machine (SVM). The evaluation results on the Kvasir-Capsule dataset show 99% accuracy for the proposed method in polyp detection. Furthermore, the proposed method runs at a real-time speed of ∼0.031 seconds detection for each image. © 2023 Bardia Delagah and Hamid Hassanpour.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Semi-Supervised Anomaly Detection in Electronic-Exam Proctoring Based on Skeleton Similarity

Semi-Supervised Anomaly Detection in Electronic-Exam Proctor...

引用

European Workshop on Visual Information processing (EUVIP)

作者： Habibollah Agh Atabay Hamid Hassanpour Image Processing & Data Mining Lab Shahrood University of Technology Shahrood Iran

关键词：

来源：评论

学校读者我要写书评

暂无评论

Abnormal Behavior Detection in Electronic-Exam Videos Using BeatGAN

Abnormal Behavior Detection in Electronic-Exam Videos Using ...

引用

International Conference of Signal processing and Intelligent Systems (ICSPIS)

作者： Habibollah Agh Atabay Hamid Hassanpour Image Processing & Data Mining Lab Shahrood University of Technology Shahrood Iran

ISBN: (纸本)9781665476249

关键词： Training Measurement Signal processing algorithms Signal processing Feature extraction data models Skeleton

来源：评论

学校读者我要写书评

暂无评论

Electronic-Exam Video Summarization Using Skeleton-Based Anomaly Detection in Time-Series data

SSRN

引用

SSRN 2022年

作者： Atabay, Habibollah Agh Hassanpour, Hamid Image Processing & Data Mining Lab Shahrood University of Technology Shahrood Iran

Recognizing cheating in electronic exams (e-exams) is a challenging issue especially when candidates hide unauthorized tools or resources from the view of proctors. Detecting abnormal behaviors is essential to recognize such events. This paper addresses the problem of abnormal behavior detection in e-exam videos as an anomaly detection task in the time-series data generated by 2D skeleton-based features. Different from the literature, our focus is on proctoring from single side-view cameras. To start, we extract skeleton-based features from each video frame of datasets, including joint coordinates, bone length, bone angle, and their combinations. The next step is training the detector models. Reconstruction-based semi-supervised deep learning models of anomaly detection are used. These models are trained on the normal instances of datasets, and reconstruction errors of the test samples are used to rank them . We select the top-ranked samples as anomalies and summarize the videos based on them. To train the models, three strategies are used. First, in the global strategy, we use the edBB dataset for training and our collected dataset for testing. The videos of the collected dataset contain the simulated cheating behaviors in the last 60% of each video. Second, in the video-specific strategy, the initial frames of each video are used for training and the remaining for testing. And third, the combined approach merges the results of the global and the video-specific strategies. As the result of anomalous-sequence detection, Univariate Auto-Encoder (UAE) recognizes 61% of anomaly sequences of the dataset in the summarized videos of the length of 10% of the original. UAE achieves this result using the combined strategy and the explicitly-extracted bone-length-and-angle features. As the result of anomalous-event detection, LSTM-VAE recognizes 93% of anomalous events using the combined strategy and the bone-length features. Looking at the problem from different aspects and

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Advancements in Content-Based image Retrieval: A Comprehensive Survey of Relevance Feedback Techniques

arXiv

引用

arXiv 2023年

作者： Qazanfari, Hamed Alyan Nezhadi, Mohammad M. Khoshdaregi, Zohreh Nozari Image Processing and Data Mining Lab Shahrood University of Technology Shahrood Iran Faculty of Science University of Science and Technology of Mazandaran Behshahr Iran Department of Computer Engineering University of Bojnord Iran

Content-based image retrieval (CBIR) systems have emerged as crucial tools in the field of computer vision, allowing for image search based on visual content rather than relying solely on metadata. This survey paper presents a comprehensive overview of CBIR, emphasizing its role in object detection and its potential to identify and retrieve visually similar images based on content features. Challenges faced by CBIR systems, including the semantic gap and scalability, are discussed, along with potential solutions. It elaborates on the semantic gap, which arises from the disparity between low-level features and high-level semantic concepts, and explores approaches to bridge this gap. One notable solution is the integration of relevance feedback (RF), empowering users to provide feedback on retrieved images and refine search results iteratively. The survey encompasses long-term and short-term learning approaches that leverage RF for enhanced CBIR accuracy and relevance. These methods focus on weight optimization and the utilization of active learning algorithms to select samples for training classifiers. Furthermore, the paper investigates machine learning techniques and the utilization of deep learning and convolutional neural networks to enhance CBIR performance. This survey paper plays a significant role in advancing the understanding of CBIR and RF techniques. It guides researchers and practitioners in comprehending existing methodologies, challenges, and potential solutions while fostering knowledge dissemination and identifying research gaps. By addressing future research directions, it sets the stage for advancements in CBIR that will enhance retrieval accuracy, usability, and effectiveness in various application domains. © 2023, CC BY.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Gait recognition based on measurements of moving human legs angles

引用

International Journal of Engineering, Transactions B: Applications 2020年第5期33卷 975-983页

作者： Al-Ani, M. Shaban Mohammadi, M. AlyanNezhadi, M.M. University of Human Development College of Science and Technology Department of Computer Science Sulaymaniyah KRG Iraq Image Processing and Data Mining Lab Shahrood University of Technology Shahrood Iran

Biometrics plays a crucial role in security systems. Gait recognition as an aspect of behavioral biometrics deals with identifying human based on the walking. In this research, we proposed a new method to identify people according to their gait characteristics. This approach is performed via several steps;a) the video frames of a walking subject is recorded. b) each video is decomposed into eight phases, wherein each phase, there are several frames. One frame per phase is selected. c) the extracted frame is segmented to separate the subject from the background. d) the image is converted to a binary image, and the skeletonization is applied to obtain the silhouette of the image. e) the fast Fourier transform is applied to each frame, and several statistical features are calculated. The average of the statistical measures for eight frames of a video is computed as a feature vector for the subject. Finally, the correlation is used to identify the subject based on the feature vector. The accuracy of 100% is achieved for the recognition of the subjects in the CASIA dataset. © 2020 Materials and Energy Research Center. All rights reserved.

关键词： Binary images

来源：评论

学校读者我要写书评

暂无评论

Content-based image retrieval considering colour difference histogram of image texture and edge orientation

引用

International Journal of Engineering, Transactions B: Applications 2020年第5期33卷 949-958页

作者： AlyanNezhadi, M.M. Qazanfari, H. Ajam, A. Amiri, Z. Image Processing and Data Mining Lab Shahrood University of Technology Shahrood Iran Department of Computer Engineering Shahrood Non-profit and Non-government Higher Edu. Inst Shahrood Iran

Content-based image retrieval is one of the interesting subjects in image processing and machine vision. In image retrieval systems, the query image is compared with images in the database to retrieve images containing similar content. image comparison is done using features extracted from the query and database images. In this paper, the features are extracted based on the human visual system. Since the human visual system considers the texture and the edge orientation in images for comparison, the colour difference histogram associated with the image’s texture and edge orientation is extracted as a feature. In this paper, the features are selected using the Shannon entropy criterion. The proposed method is tested using the Corel-5K and Corel-10K databases. The precision and recall criteria were used to evaluate the proposed system. The experimental results show the ability of the proposed system for more accurate retrieval rather than recently content-based image retrieval systems. © 2020 Materials and Energy Research Center. All rights reserved.

关键词： data mining

来源：评论

学校读者我要写书评

暂无评论

Interval probabilistic fuzzy WordNet

arXiv

引用

arXiv 2021年

作者： Alizadeh-Q, Yousef Minaei-Bidgoli, Behrouz Hossayni, Sayyed-Ali Akbarzadeh-T, Mohammad-R. Recupero, Diego Reforgiato Rajati, Mohammad-Reza Gangemi, Aldo Data Mining Lab School of Computer Engineering Iran University of Science and Technology Tehran Iran Department of Computer Engineering Ferdowsi University of Mashhad Mashhad Iran Agents Research Lab TECNIO Centre EASY ViCOROB Research Institute Campus de Montilivi University of Girona Girona Catalonia Spain Italy Laboratoire d'Informatique de Paris Nord Université Paris 13 Sorbonne Paris Cité-CNRS France Signal and Image Processing Institute University of Southern California Los AngelesCA90089-02564 United States

WordNet lexical-database groups English words into sets of synonyms called "synsets." Synsets are utilized for several applications in the field of text-mining. However, they were also open to criticism because although, in reality, not all the members of a synset represent the meaning of that synset with the same degree, in practice, they are considered as members of the synset, identically. Thus, the fuzzy version of synsets, called fuzzy-synsets (or fuzzy word-sense classes) were proposed and studied. In this study, we discuss why (type-1) fuzzy synsets (T1 F-synsets) do not properly model the membership uncertainty, and propose an upgraded version of fuzzy synsets in which membership degrees of word-senses are represented by intervals, similar to what in Interval Type 2 Fuzzy Sets (IT2 FS) and discuss that IT2 FS theoretical framework is insufficient for analysis and design of such synsets, and propose a new concept, called Interval Probabilistic Fuzzy (IPF) sets. Then we present an algorithm for constructing the IPF synsets in any language, given a corpus and a word-sense-disambiguation system. Utilizing our algorithm and the open-American-online-corpus (OANC) and UKB word-sense-disambiguation, we constructed and published the IPF synsets of WordNet for English language. Copyright © 2021, The Authors. All rights reserved.

关键词： Ontology

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：