检索结果-内蒙古大学图书馆

Memory-enhanced spatial-temporal encoding framework for industrial anomaly detection system

EXPERT SYSTEMS WITH APPLICATIONS 2024年 250卷

作者： Liu, Yang Ju, Bobo Yang, Dingkang Peng, Liyuan Li, Di Sun, Peng Li, Chengfang Yang, Hao Liu, Jing Song, Liang Fudan Univ Acad Engn & Technol Shanghai 200433 Peoples R China Univ Toronto Dept Comp Sci Toronto ON M5S 1A1 Canada Duke Kunshan Univ Div Nat & Appl Sci Kunshan 215316 Jiangsu Peoples R China Fudan Univ State Key Lab ASIC & Syst Shanghai 201203 Peoples R China Ningbo Univ Fac Elect Engn & Comp Sci Ningbo 315211 Zhejiang Peoples R China Fudan Univ Sch Informat Sci & Technol Shanghai 200433 Peoples R China Univ British Columbia Dept Elect & Comp Engn Vancouver BC V6T 1Z4 Canada Innovat Platform Academicians Hainan Prov Haikou 570228 Hainan Peoples R China

The development of modern manufacturing has raised greater demands on the accuracy, response speed, and operating cost of industrial accident warnings. Compared to conventional contact sensors, surveillance cameras can contactlessly capture spatial-temporal information of the open workspace with stable data quality, widely used in industrial process monitoring. However, due to the scarcity of industrial video datasets and the rarity and diversity of abnormal events, existing video -based anomaly detection models perform poorly in manufacturing scenarios. In this regard, we collect two datasets from typical industrial sites and propose a memory -enhanced spatial-temporal encoding (MSTE) framework for automatic industrial anomaly detection. The proposed MSTE framework learns spatial and temporal normality as well as spatial-temporal correlations with parallel structures and simultaneously measures deviations in appearance, motion, and consistency to respond to complex industrial anomalies accurately. Experimental results on public benchmarks and realworld industrial videos show that our method outperforms existing methods and achieves accurate temporal localization of various spatial-temporal anomalies, which helps to improve the safety and reliability of intelligent manufacturing.

关键词： Industrial anomaly detection Intelligent information system Spatial-temporal anomaly detection deep autoencoder Normality learning

来源：评论

学校读者我要写书评

暂无评论

Optimizing machine learning algorithms for fault classification in rolling bearings: A Bayesian Optimization approach

引用

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2025年 150卷

作者： Yousaf, Muhammad Zain Guerrero, Josep M. Sadiq, Muhammad Tariq Zhejiang Univ Ctr Renewable Energy & Microgrids Huanjiang Lab Zhuji 311816 Zhejiang Peoples R China Univ Valladolid Dept Elect Engn Valladolid 47011 Spain Aalborg Univ AAU Energy Ctr Res Microgrids CROM DK-9220 Aalborg Denmark Univ Essex Sch Comp Sci & Elect Engn Colchester CO4 3SQ Essex England

Modern power machinery is inherently complex and operates under dynamic operating conditions, so they demand advanced solutions based on deep learning to diagnose bearing faults inside rotating equipment that cause unplanned downtime and safety issues, leading to operational challenges. However, most deep learning approaches aim to improve performance by incorporating hybrid neural networks that rely on multiple convolutional and temporal units, often overlooking optimizing the large number of hyperparameters that define the structure and performance of hybrid models along with the associated computational constraints. To address this gap, this study presents an innovative approach for the detection and classification of bearing faults by integrating an optimized sparse deep autoencoder (DAE) with a Bidirectional Long Short-Term Memory model (BiLSTM). The optimal network structure and hyperparameters are determined through Bayesian optimization (BO) with parallel settings, which automatically searches for network configurations that improve the feature extraction ability of the DAE and the generalization ability of the Bi-LSTM for more efficient fault classification in rolling bearings. Parallel optimization accelerates network structure and hyperparameter tuning by evaluating multiple configurations at once. It leverages the full potential of available multi-core Central Processing Units (CPUs)/Graphics Processing Units (GPUs) in conjunction with a lightweight BO surrogate model. This autonomous and user-friendly framework generates inputs from principal component analysis for linear and BO-DAE for non-linear feature extraction and selection, which are then used to train a BO-enhanced Bi-LSTM. This threestage optimized method effectively captures spatial and temporal dependencies in vibrational signals, achieving superior efficiency, accuracy, and reliability compared to shallow and deep learning models. Evaluation metrics, including macro precision (99.50 %), reca

关键词： Rolling element bearings Modern power machinery deep autoencoder Bidirectional long short-term memory Bayesian optimization

来源：评论

学校读者我要写书评

暂无评论

RAE-SVM based Webpage Tamper-Resistant Detection Algorithm

引用

JOURNAL OF INFORMATION PROCESSING SYSTEMS 2025年第1期21卷 1-12页

作者： Zhou, Changjian Zhang, Yutong Liang, Yunfu Xing, Jinge Northeast Agr Univ Dept Modern Educ Technol Harbin Peoples R China Northeast Agr Univ Coll Elect & Informat Harbin Peoples R China

Cybersecurity has become a key component of national strategy in recent years. Traditional cybersecurity technology such as network traffic-based intrusion detection and threatening intelligence sensing are designed to focus on the traffic features of network, which are no doubt effective defense technologies. However, these methods required decent amount of domain knowledge and massive training data, which brought a significant barrier for cybersecurity research. In this work, we propose a novel residual autoencoder and support vector machine combined approach (RAE-SVM) for webpage tamper-resistant detection using high-level webpage image features. This method, inspired by the Chinese proverb "mend the fold after the sheep have been stolen." The web crawler technology is used for website screenshot within limited domain names, and input them into autoencoder architecture and SVM for feature extraction and invaded webpage detection. This method combines the advantages of deep residual network, convolutional autoencoder and SVM, and the interdisciplinary intersection between cybersecurity and high-level image features. The experimental results demonstrate that the proposed method achieves an accuracy of 95%, significantly higher than other models, which proves the validity of the proposed method.

关键词： Cyber Security deep autoencoder Image Features Support Vector Machine Tamper-Resistant Detection

来源：评论

学校读者我要写书评

暂无评论

Hyperparameter optimization in unsupervised anomaly detection for mineral prospectivity mapping

引用

ORE GEOLOGY REVIEWS 2025年 181卷

作者： Mirzabozorg, Seyyed Ataollah Agha Seyyed Saremi, Mobin DehghanNiri, Ramin Abedi, Maysam Yousefi, Mahyar Pour, Amin Beiranvand Hezarkhani, Ardeshir Maghsoudi, Abbas Univ Tehran Coll Engn Sch Min Engn Tehran Iran Amirkabir Univ Technol Dept Min Engn Tehran Iran Malayer Univ Fac Engn Dept Min Engn Malayer Iran

Mineral potential mapping (MPM) can recognise irregular patterns of mineralization-related indicator features and proxies. It serves as an anomaly detection technique, given that mineralization itself is a rare geological event. In this regard, unsupervised anomaly detection (UAD) algorithms could be effective in identifying high potential zones of mineralization accounting for irregular pattern recognition. The main advantage of these algorithms lies in their ability toexploit geo-datasets without requiring any form of annotation. In this study, eight evidence layers were first created based on the conceptual model of mineral deposits to build a model of Fe prospectivity in the Esfordi region of Yazd province, located in east-central Iran. Then, three unsupervised anomaly detection algorithms, namely deep autoencoder (DAE), one-class support vector machine (OC-SVM), and isolation forest (IForest) were employed to assess Fe prospectivity in the area. The prediction-area (P-A) plot was subsequently used to evaluate the efficacy of the three prospectivity models. Finding indicate that the deep autoencoder outperforms the other adopted machine learning methods in identifying high potential areas of Fe mineralization. Considering the significance of hyperparameters in the implementation of these algorithms, we also investigate the application of the P-A plot to identify optimal hyperparameter values, thereby enhancing the performance of the Fe prospectivity model. The results demonstrate that in IForest and DAE, and to some extent OC-SVM, experts can adjust hyperparameters without relying on labelled data, achieving a commendable level of detection performance. This innovative approach and workflow are broadly applicable to regional-scale mineral exploration across diverse metallogenic provinces globally.

关键词： Mineral potential mapping Unsupervised anomaly detection deep autoencoder Fe prospecting model Esfordi district Iran

来源：评论

学校读者我要写书评

暂无评论

An SSA-LC-DAE Method for Extracting Network Security Elements

引用

IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING 2023年第2期10卷 1175-1185页

作者： Tao, Xiaoling Liu, Ziyi Zhao, Feng Lan, Rushi Liu, Runrong Fu, Lianyou Ouyang, Yifu Guilin Univ Elect Technol Sch Comp Sci & Informat Secur Guilin 541004 Peoples R China Guilin Univ Elect Technol Sch Informat & Commun Guilin 541004 Peoples R China

In the current network environment, the original network data present the characteristics of multiple features and high dimensions. Furthermore, in the existing element extraction methods based on deep neural networks, the loss of feature information layer by layer increases as the data dimensions decrease, which greatly affects the retention of network data information elements and brings a huge challenge to effective network security protection. This paper refers to a residual neural network to improve the deep autoencoder (DAE) and then utilizes them to propose a novel element extraction method named the layer-by-layer loss compensation deep autoencoder (LC-DAE) based on the sparrow search algorithm (SSA). In the proposed method, a loss compensation module is added to each encoding layer of the DAE. Specifically, this module first restores the data by using the decoding layer corresponding to the encoding layer. Then, the loss value of the calculated characteristic information is compensated using the output of the corresponding encoding layer. Subsequently, the SSA is used to optimize the LC-DAE in the training process. Finally, the experimental results show that compared with the existing methods, this method retains more sufficient element information, and significantly improves the classification performance of data using neural networks.

关键词： Feature extraction Data mining Network security Encoding Decoding Neural networks deep learning deep autoencoder element extraction Index Terms network data sparrow search algorithm

来源：评论

学校读者我要写书评

暂无评论

A Reminiscent Intrusion Detection Model Based on deep autoencoders and Transfer Learning

A Reminiscent Intrusion Detection Model Based on Deep Autoen...

引用

IEEE Global Communications Conference (GLOBECOM)

作者： dos Santos, Roger R. Viegas, Eduardo K. Santin, Altair O. Pontificia Univ Catolica Parana Grad Program Comp Sci PPGIa Curitiba Parana Brazil

ISBN: (纸本)9781728181042

Machine learning techniques for network-based intrusion detection often assume that network traffic does not change over time or that model updates can be easily performed. This paper proposes a novel, reminiscent intrusion detection model based on deep autoencoders and transfer learning to ease the model update burden in a twofold implementation. First, a deep autoencoder is used as an additional feature extraction stage to obtain a historical feature representation of network traffic. Second, at model updates, the deep autoencoder parameters are updated through a transfer learning procedure, thus, significantly decreasing the amount of needed labeled training data and the computational costs. Experiments performed on a 8TB dataset containing real and valid network traffic ranging for one year have shown that approaches in the literature cannot handle with the network traffic behavior changes over time, requiring impractical amounts of labeled data to be provided during model training tasks. In addition, if no model updates are performed, the proposed scheme can improve the true-negative rate by up to 23.9%. If done so, it can provide similar accuracy rates of traditional techniques while demanding only 22% of labeled training data and 28% of computational costs.

关键词： Intrusion Detection deep autoencoder Transfer Learning

来源：评论

学校读者我要写书评

暂无评论

Dynamic Viewing Pattern Analysis: Towards Large-Scale Screening of Children With ASD in Remote Areas

引用

IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING 2023年第5期70卷 1622-1633页

作者： Xia, Chen Zhang, Dingwen Li, Kuan Li, Hongxia Chen, Jianxin Min, Weidong Han, Junwei Northwestern Polytech Univ Sch Automat Xian Peoples R China Northwest Univ Children Healthcare Dept Affiliated Hosp Xian Peoples R China Shaanxi Rehabil Hosp Children Rehabil Ctr Xian Peoples R China Beijing Univ Chinese Med Sch Tradit Chinese Med Beijing Peoples R China Nanchang Univ Sch Math & Comp Sci Nanchang Peoples R China Northwestern Polytech Univ Sch Automat Xian 710072 Peoples R China Hefei Comprehens Natl Sci Ctr Inst Artificial Intelligence Hefei 230088 Peoples R China

Objective: Autism spectrum disorder (ASD) affects nearly 1 in 44 children younger than 8 years old in the United States, and the situation may be even worse in remote areas of the world. However, it is difficult to utilize existing approaches to screen patients with ASD in remote areas due to the lack of professionals and high-tech instruments. Therefore, we develop a fast and accurate scalable method for screening children with ASD. Methods: A deep weakly supervised artificial intelligence model is proposed for ASD screening based on the dynamic viewing patterns (DVP) over viewing time and visual stimuli. In training, we utilized a long short-term memory (LSTM) network to learn the mapping between the autoencoder-based encoded dynamic patterns and the labels. In testing, we fed the encoded DVP of each undiagnosed child into the trained network and predicted the diagnosis category based on the score on all stimuli. Results: Based on the multi-center evaluation on 165 subjects (95 typically developing children and 70 children with ASD) aged 3-6 years from different areas of China, our method achieves an average recognition accuracy of 96.73% (sensitivity 96.85% and specificity 96.63%). Conclusion: The DVP is a discriminating attribute to identify the atypical performance of ASD. The DVP-based model is an effective platform for enhancing auxiliary ASD screening accuracy. Significance: We validated the importance of dynamic information on between-group differences and classification. Additionally, the evaluation results suggest that the proposed model can provide an objective and accessible tool for scalable ASD screening applications.

关键词： Autism spectrum disorder deep autoencoder deep learning dynamic viewing patterns long short-term memory (LSTM) network

来源：评论

学校读者我要写书评

暂无评论

Hybrid Attention and Motion Constraint for Anomaly Detection in Crowded Scenes

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 2023年第5期33卷 2259-2274页

作者： Zhang, Xinfeng Fang, Jinpeng Yang, Baoqing Chen, Shuhan Li, Bin Yangzhou Univ Coll Artificial Intelligence Coll Informat Engn Jiangsu Prov Engn Res Ctr Knowledge Management Yangzhou 225127 Jiangsu Peoples R China

Crowds often appear in surveillance videos in public places, from which anomaly detection is of great importance to public safety. Since the abnormal cases are rare, variable and unpredictable, autoencoders with encoder and decoder structures using only normal samples have become a hot topic among various approaches for anomaly detection. However, since autoencoders have excessive generalization ability, they can sometimes still reconstruct abnormal cases very well. Recently, some researchers construct memory modules under normal conditions and use these normal memory items to reconstruct test samples during inference to increase the reconstruction errors for anomalies. However, in practice, the errors of reconstructing normal samples with the memory items often increase as well, which makes it still difficult to distinguish between normal and abnormal cases. In addition, the memory-based autoencoder is usually available only in the specific scene where the memory module is constructed and almost loses the prospect of cross-scene applications. We mitigate the overgeneralization of autoencoders from a different perspective, namely, by reducing the prediction errors for normal cases rather than increasing the prediction errors for abnormal cases. To this end, we propose an autoencoder based on hybrid attention and motion constraint for anomaly detection. The hybrid attention includes the channel attention used in the encoding process and spatial attention added to the skip connection between the encoder and decoder. The hybrid attention is introduced to reduce the weight of the feature channels and regions representing the background in the feature matrix, which makes the autoencoder features more focused on optimizing the representation of the normal targets during training. Furthermore, we introduce motion constraint to improve the autoencoder's ability to predict normal activities in crowded scenes. We conduct experiments on real-world surveillance videos, UCSD, CU

关键词： Videos Anomaly detection Training Memory modules Dictionaries Testing Surveillance video surveillance deep autoencoder attention mechanism

来源：评论

学校读者我要写书评

暂无评论

deep discriminative dictionary pair learning for image classification

引用

APPLIED INTELLIGENCE 2023年第19期53卷 22017-22030页

作者： Zhu, Wenjie Peng, Bo Chen, Chunchun Chen, Hao China Jiliang Univ Coll Informat Engn Key Lab Electromagnet Wave Informat Technol & Metr Hangzhou 310018 Zhejiang Peoples R China Univ Queensland St Lucia Brisbane Qld 4072 Australia China Jiliang Univ Coll Sci Hangzhou 310018 Zhejiang Peoples R China

Discriminative dictionary learning has been extensively used for pattern classification tasks. By incorporating different kinds of label information into the dictionary learning framework, a dictionary can be attained that represents the original signal with discriminative reconstruction. The previous works learn the dictionary in the original space which limits the dictionary learning performance. In this paper, we propose an approach, namely deep Discriminative Dictionary Pair Learning ((DPL)-P-3) for image classification. The input of (DPL)-P-3 is not the matrix collected by original gray images or hand-crafted features but the relatively deeper features derived from autoencoders. Then, a structured dictionary is designed based on the discriminative contributions across different classes to reconstruct the deep feature. In addition, the associated structured projective dictionary is learned as well to guarantee the decoders updating towards the minimal error of deconvolution operator. By leveraging the discriminative-dictionary-learning-based loss function and the autoencoder loss function, (DPL)-P-3 can simultaneously learn the deep potential feature and the corresponding dictionary pair. In the testing phase of (DPL)-P-3, the minimum error between the deep feature and the structured projective component with regard to different classes can directly indicate the label by a basic matrix multiplication operation. Experimental results on challenging Extended Yale B, AR, UMIST, COIL20, Scene 15, and Caltech101 datasets demonstrate that the proposed (DPL)-P-3 outperforms the prominent dictionary learning methods.

关键词： Dictionary learning deep autoencoder Projective dictionary Dictionary pair learning Image classification

来源：评论

学校读者我要写书评

暂无评论

deep auto encoder based on a transient search capsule network for student performance prediction

引用

MULTIMEDIA TOOLS AND APPLICATIONS 2023年第15期82卷 23427-23451页

作者： Rahul Katarya, Rahul Delhi Technol Univ Dept Comp Sci & Engn Big Data Analyt & Web Intelligence Lab New Delhi India

Prediction of Student performance through a machine predicts a student's future success. It can be considered an essential procedure to determine the students' academic excellence and identify them at high risk for academic performance. Prediction of student performance also provides universities with a high reputation and ranking. The evaluation of 'What students can do with their learning' is still a tedious task. There are many challenging factors to solve this problem, mainly owing to the enormous amount of data collected from students. Most of the research works have focused on developing new methodologies for student performance prediction. But all the existing work has some performance limitations. Here, a new model called transient search capsule network based on the deep autoencoder (TSCNDE) is introduced to detect student performance. The TSCNDE method is implemented with the help of the PYTHON tool. The performance prediction process has been completed with the help of the OULA dataset. The obtained results are assessed on accuracy (99.2%), precision, (99.8%), specificity (98.7%), and sensitivity (98.9%) parameters. The results obtained showed that the TSCNDE method is about 99.2% more accurate than the other related method. Also, the obtained results are compared with some existing deep learning and machine learning methods.

关键词： deep learning Student performance prediction OULA dataset Classification deep autoencoder Capsule network

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

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

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：