检索结果-内蒙古大学图书馆

8th European Workshop on Visual Information Processing (EUVIP)

作者： Boutarfass, Sanae Besserer, Bernard Univ La Rochelle MIA Lab La Rochelle France

ISBN: (纸本)9781728144962

We describe in this paper how we use a CNN-autoencoder for unsupervised anomaly detection, here adapted to detect anomalies in handwritten text samples. There is no distinction between a learning and test dataset, the text snippet containing the anomaly is the only data available. Since character recognition is not aimed at, we expect that the autoencoder should learn the styling characteristics of the writing. Algorithms used for graphology commonly measure the extends of the vertical strokes over or under the base line, the slope of the strokes or the closure of letters such as the "o". Given the text baseline which is detected by image processing, the image of the sample is arbitrarily split in tiles for the training / recognition task. We also use a partial Radon projection to convert these tiles in a more abstracted representation for the learning/detection tasks. The Radon projection is computed over a [pi/2] range which captures the most style variations in handwritten text. Discrimination takes place directly while training, and by shifting image tiles from batch to batch as the training epochs go on, outliers are discarded and localised.

关键词： deep learning handwriting outlier detection convolutional autoencoder radon transform

来源：评论

学校读者我要写书评

暂无评论

convolutional autoencoder-based Sensor Fault Classification 10

Convolutional Autoencoder-based Sensor Fault Classification

引用

10th International Conference on Ubiquitous and Future Networks (ICUFN)

作者： Yang, Jae-Wan Lee, Young-Doo Koo, In-Soo Univ Ulsan Dept Elect Engn Ulsan South Korea

ISBN: (纸本)9781538646465

Automation machines perform not only simple operations but also operations requiring high accuracy. Sensors are essential to carry out the delicate operations. Therefore, if there is a fault in sensors, the machine can malfunction and the process-line will be damaged. To prevent this, sensors should be monitored and diagnosed in real time. In the paper, we propose a convolutional autoencoder-based sensor fault classification scheme in which time-domain statistical features and convolutional autoencoder features of sensor data are both utilized to classify types of sensor faults. Through simulation, it is shown that the proposed scheme can improve classification performance of the sensor faults.

关键词： Sensor faults Fault detection Time-domain statistical feature convolutional autoencoder Support vector machine Erratic fault Drift fault Hard-over fault Spike fault Stuck fault

来源：评论

学校读者我要写书评

暂无评论

convolutional autoencoder for Compressive Symbol Detection 9

Convolutional Autoencoder for Compressive Symbol Detection

引用

9th International Conference on Information and Communication Technology Convergence (ICTC)

作者： Park, Jae-Hyuck Kim, Yookyung Elect & Telecommun Res Inst Broadcasting & Media Res Lab Daejeon South Korea

ISBN: (纸本)9781538650417

This paper deals with a problem of digital communication which samples a signal compressively below its symbol rate and restores the original message symbols. We propose a method to optimize the sensing and restoration modules by designing them as the encoder and decoder of a convolutional autoencoder. The network learns end -to-end through simulation data. Our architecture has an efficient mechanism to solve this compressive symbol detection problem and it shows strong performance and fast speed empirically.

关键词： Compressive Symbol Detection convolutional autoencoder

来源：评论

学校读者我要写书评

暂无评论

Multi-bearing fault diagnosis method based on convolutional autoencoder causal decoupling domain generalization

引用

ISA Transactions 2025年

作者： Xinyang Cui Hongfei Zhan Kang Han Junhe Yu Rui Wang Guojun Huang Faculty of Mechanical Engineering and Mechanics Ningbo University Ningbo 315211 China Ningbo JinYu Technology Industry Co. Ltd Ningbo315400 China

Currently, most transfer learning (TL) methods for solving bearing fault diagnosis primarily focus on diagnosing a single bearing. However, in practical engineering scenarios, multi-bearing co-occurrence is often encountered, and joint diagnostic research under such conditions has received less attention. Additionally, it is challenging to collect sufficient fault samples for training, and due to changes in working conditions, the collected signals often follow different distributions. These factors limit the generalization ability of transfer learning models. Furthermore, most fault diagnosis methods for domain generalization primarily model the statistical correlation between time-series data and labels without effectively predicting the actual data, resulting in a relatively superficial description of the data generation process, which may introduce personal bias. To address these challenges, this paper proposes a causal decoupling network based on convolutional autoencoder (CDN-CAE) to diagnose and locate faults in multi-bearing systems using vibration signals from auxiliary bearings. First, maximum entropy optimization and mutual information neural estimation are introduced to decouple the time-series data into causal and non-causal factors, thereby reconstructing the data generation process comprehensively. On this basis, an aggregation loss is proposed to separate causal and non-causal factors, enabling better aggregation of the same fault type and improving the generalization ability of the method. Finally, a reconstruction loss is introduced to ensure the completeness of the information within the decoupled factors and to enhance the network's robustness. Experiments conducted on a multi-bearing device demonstrate that the proposed method achieves high accuracy and overall stability.

关键词： Causal learning convolutional autoencoder Domain generalization Multi-bearing fault diagnosis

来源：评论

学校读者我要写书评

暂无评论

convolutional autoencoder based Feature Extraction in Radar Data Analysis 10

Convolutional Autoencoder based Feature Extraction in Radar ...

引用

Joint 10th International Conference on Soft Computing and Intelligent Systems (SCIS) / 19th International Symposium on Advanced Intelligent Systems (ISIS)

作者： Lee, Hansoo Kim, Jonggeun Kim, Baekcheon Kim, Sungshin Pusan Natl Univ Dept Elect & Comp Engn Busan 46241 South Korea

ISBN: (纸本)9781538626337

One of the challenging research topics is a fast, accurate, and human-like image processing method. The convolutional neural network recently shows optimistic results, but it needs a vast amount of computational resources. Utilizing feature extraction and dimensionality reduction algorithm might be a solution to increase computational efficiency. In this paper, we implemented a convolutional autoencoder for performing feature extraction and dimensionality reduction which not only can solve nonlinear problems but also easily combine the convolutional neural network. The implemented convolutional autoencoder derives remarkable reconstruction error in the experiments using the two-dimensional radar data.

关键词： convolutional autoencoder feature extraction radar images

来源：评论

学校读者我要写书评

暂无评论

Deep Clustering Network for Steganographer Detection Using Latent Features Extracted from a Novel convolutional autoencoder

引用

NEURAL PROCESSING LETTERS 2023年第3期55卷 2953-2964页

作者： Amrutha, E. Arivazhagan, S. Jebarani, W. Sylvia Lilly Mepco Schlenk Engn Coll Autonomous Ctr Image Proc & Pattern Recognit Dept Elect & Commun Engn Sivakasi 626005 Tamil Nadu India

Steganography is typically used by law enforcement agencies to prevent unauthorized persons from becoming aware of the existence of a message communicated by military or other government organizations outside their network. Illegal uses of steganography such as fraud, gambling, criminal communications, hacking, electronic payments, harassment, offenses on intellectual property and viruses pose a great threat to society. Such illicit steganographers communicate with each other through stego files to exchange their plan without getting noticed by law enforcement. This paper presents a novel Deep Clustering Network for Steganographer Detection (DCNSD) based on convolutional autoencoders and a clustering model for identifying such steganographers since plenty of digital images are transferred over the internet that could carry hidden secret messages. Mostly the existing techniques involving deep learning networks for steganographer detection involves supervised learning approach which makes them unsuitable for real-world deployment. The foremost characteristic of this proposed network lies in its ability to segregate images transmitted from a steganographer from those innocent users' images in an unsupervised approach. Thereby making the proposed DCNSD system plausible for real-world deployment. Based on the experimental results, it is shown that the proposed DCNSD framework excels in detecting steganographers who use content-adaptive steganography to embed secrets with 100% accuracy.

关键词： Steganographer detection convolutional autoencoder Latent feature Deep clustering Cluster evaluation metrics

来源：评论

学校读者我要写书评

暂无评论

Condition-invariant and compact visual place description by convolutional autoencoder

引用

ROBOTICA 2023年第6期41卷 1718-1732页

作者： Ye, Hanjing Chen, Weinan Yu, Jingwen He, Li Guan, Yisheng Zhang, Hong Southern Univ Sci & Technol Shenzhen Key Lab Robot & Comp Vis Shenzhen Peoples R China Southern Univ Sci & Technol Dept Elect & Elect Engn Shenzhen Peoples R China Guangdong Univ Technol Sch Mech & Elect Engn Guangzhou Peoples R China

Visual place recognition (VPR) in condition-varying environments is still an open problem. Popular solutions are convolutional neural network (CNN)-based image descriptors, which have been shown to outperform traditional image descriptors based on hand-crafted visual features. However, there are two drawbacks of current CNN-based descriptors: (a) their high dimension and (b) lack of generalization, leading to low efficiency and poor performance in real robotic applications. In this paper, we propose to use a convolutional autoencoder (CAE) to tackle this problem. We employ a high-level layer of a pre-trained CNN to generate features and train a CAE to map the features to a low-dimensional space to improve the condition invariance property of the descriptor and reduce its dimension at the same time. We verify our method in four challenging real-world datasets involving significant illumination changes, and our method is shown to be superior to the state-of-the-art. The code of our work is publicly available at https://***/MedlarTea/CAE-VPR.

关键词： visual place recognition dimension reduction convolutional autoencoder visual navigation

来源：评论

学校读者我要写书评

暂无评论

Enhancing representational learning for cloud robotic vision through explainable fuzzy convolutional autoencoder framework

引用

SOFT COMPUTING 2023年 1-11页

作者： Udendhran, R. Yamini, G. Badrinath, N. Amalraj, J. Jegathesh Suresh, A. SRM Inst Sci & Technol Sch Comp Dept Computat Intelligence Chegalpattu 603203 India Natl Coll Autonomous PG & Res Dept Comp Sci Tiruchirappalli 620001 India Vellore Inst Technol Sch Comp Sci & Engn Vellore 632014 India Govt Arts & Sci Coll Dept Comp Sci Cuddalore 606106 India SRM Inst Sci & Technol Fac Engn & Technol Sch Comp Dept Networking & Commun Chengalpattu 603203 India

Robot vision is one of the most recent developments in robotics and automation technology. Owing to machine vision systems that integrate image processing and deep learning, robots could now operate faster on the assembly line and in new contexts such as supermarkets, hospitals, and restaurants. The key driver for such systems is the advancement of machine vision systems. Robotic vision is mainly composed of programs, cameras, and any other technology that assists robots in developing visual insights. This enables robots to do sophisticated visual tasks, such as a robot arm trained to pick up an object. The conventional algorithms employ semantic segmentation from camera images. Reconstructing a real-time environment from two or more images entails creating a 3D model of the scene require expensive labeled data to train visual semantic segmentation. These 3D models can be anything from fragments of a 3D point cloud to 3D surface models produced using advanced techniques. In order to extract several images into 3D models or subsets of a raw point cloud, enhanced representation learning is necessary. Inspired by the recent achievements of machine vision and deep learning, this paper proposes a deep representation learning technique for enhanced feature recognition learning based on explainable convolutional autoencoder that can be employed for feature classifier level using fuzzy clustering and further facilitated the human decision making of data, which requires for the representation of robot data that allow for swift and precise observations.

关键词： Deep learning convolutional autoencoder Robot vision Feature recognition 3D modeling Robotics Computer vision

来源：评论

学校读者我要写书评

暂无评论

A video system based on convolutional autoencoder for drowning detection

引用

NEURAL COMPUTING & APPLICATIONS 2023年第21期35卷 15791-15803页

作者： He, Xinyu Yuan, Fei Liu, Tingzhuang Zhu, Yi Xiamen Univ Sch Informat Xiamen Peoples R China

Computer vision combined with deep learning technologies is widely used in video surveillance. In this paper, it is applied to drowning detection video systems. Traditional drowning detection methods detect drowning mainly by monitoring the physiological condition, the time and motion of swimmers in the water. But these methods are not applicable to detect early quiet drowning phenomena. Some researchers realize supervised classification by simulative drowning features. But real drowning events are difficult to truly simulate, so these methods are not reliable. In this paper, a drowning detection video system with edge computing is proposed, and it can detect drowning events in swimming pools without any wearable devices. According to the characteristics of drowning people, the strategies for underwater near-vertical human detection are proposed, providing a reliable basis for drowning detection. A lightweight drowning detection convolutional autoencoder is proposed to achieve unsupervised drowning detection, solving the lack of drowning videos and the inauthenticity of simulative videos. Then, an edge device is designed for detecting drowning in real time at the edge. Finally, for training and experimental evaluation, a pool dataset including many pool underwater video sequences is produced. The experimental results show that the proposed drowning detection method has a good comprehensive performance. The system is feasible and valuable.

关键词： Drowning detection Underwater videos convolutional autoencoder Unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

Global and Local Feature Extraction Using a convolutional autoencoder and Neural Networks for Diagnosing Centrifugal Pump Mechanical Faults

引用

IEEE ACCESS 2021年 9卷 65838-65854页

作者： Prosvirin, Alexander E. Ahmad, Zahoor Kim, Jong-Myon Univ Ulsan Sch Elect Elect & Comp Engn Ulsan 44610 South Korea

Centrifugal pumps are important types of electro-mechanical machines used for fluid and energy conveyance. Mechanical faults in centrifugal pumps lead to abnormal impacts in the vibration signal of the system. Those impacts induce nonstationarity in vibration signals and hence complex time-frequency domain signal analysis techniques are required to investigate the mechanical fault features of centrifugal pumps. In this paper, an end-to-end pipeline for diagnosing faults in centrifugal pumps is proposed. To create a two-dimensional representation of the transients that appear in the vibration signals due to centrifugal pump operating conditions, first, a 1/3-binary tree fast kurtogram is computed. Next, a convolutional autoencoder and convolutional neural network are trained to autonomously extract global and local features from the kurtograms. Then, global, and local features are merged to form a joined feature vector that contains different visual features that are extracted using convolutional deep architectures using their specific loss functions during the training. Finally, this feature vector is propagated to a shallow-structured artificial neural network to accomplish fault identification. The proposed framework has been validated by the dataset collected from a real industrial centrifugal pump test rig. The results obtained during the series of experimental trials demonstrated that the introduced method achieved high classification accuracies when diagnosing faults based on signals collected under 3.0 and 4.0 bars of pressure.

关键词： Feature extraction Vibrations Fault diagnosis Seals Pumps Data mining Convolution Artificial neural network centrifugal pump convolutional autoencoder convolutional neural network deep learning fault diagnosis kurtogram vibration signals

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：