检索结果-内蒙古大学图书馆

CNN adversarial attack mitigation using perturbed samples training

MULTIMEDIA TOOLS AND APPLICATIONS 2021年第14期80卷 22077-22095页

作者： Hashemi, Atiye Sadat Mozaffari, Saeed Semnan Univ Fac Elect & Comp Engn Semnan Iran

Susceptibility to adversarial examples is one of the major concerns in convolutional neural networks (CNNs) applications. Training the model with adversarial examples, known as adversarial training, is a common countermeasure to tackle such attacks. In reality, however, defenders are uninformed about how adversarial examples are generated by the attacker. Therefore, it is pivotal to utilize more general alternatives to intrinsically improve the robustness of models. For this purpose, we train CNNs with perturbed samples manipulated by various transformations and contaminated by different noises to foster robustness of networks against adversarial attacks. This idea derived from the fact that both adversarial and noisy samples undermine the classifier accuracy. We propose combination of a convolutional denoising autoencoder with a classifier (CDAEC) as a defensive structure. The proposed method does not add to the computational cost. Experimental results on MNIST database demonstrate that the accuracy of CDAEC trained by perturbed samples against adversarial attacks was more than 71.29%.

关键词： Adversarial example Convolution neural network denoising autoencoder Evasion attacks Noisy training

来源：评论

学校读者我要写书评

暂无评论

Speech Enhancement Based on Analysis-Synthesis Framework with Improved Parameter Domain Enhancement

引用

JOURNAL OF SIGNAL PROCESSING SYSTEMS FOR SIGNAL IMAGE AND VIDEO TECHNOLOGY 2016年第2期82卷 141-150页

作者： Liu, Bin Tao, Jianhua Wen, Zhengqi Mo, Fuyuan Chinese Acad Sci Inst Automat Natl Lab Pattern Recognit Beijing 100190 Peoples R China Chinese Acad Sci Inst Acoust Beijing 100190 Peoples R China

This paper presents a speech enhancement approach based on analysis-synthesis framework. An improved multi-band summary correlogram (MBSC) algorithm is proposed for pitch estimation and voiced/unvoiced (V/UV) detection. The proposed pitch detection algorithm achieves a lower pitch detection error compared with the reference algorithm. The denoising autoencoder (DAE) is applied to enhance the line spectrum frequencies (LSFs). The reconstruction loss could be decreased compare with the swallow model. The proposed approach is evaluated using the perceptual evaluation of speech quality (PESQ) and the experimental results show that the proposed approach improves the performance of speech enhancement compared with the conventional speech enhancement approach. In addition, it could be applied to parametric speech coding even at low bit rate and low signal-noise ratio (SNR) environments.

关键词： Analysis-synthesis framework Multi-band summary correlogram denoising autoencoder Speech enhancement Speech coding

来源：评论

学校读者我要写书评

暂无评论

Intelligent Identification of Simultaneous Faults of Automotive Software Systems Under Noisy and Imbalanced Data Using Ensemble LSTM and Random Forest

引用

IEEE ACCESS 2023年 11卷 140022-140040页

作者： Abboush, Mohammad Knieke, Christoph Rausch, Andreas Tech Univ Clausthal Inst Software & Syst Engn D-38678 Clausthal Zellerfeld Germany

According to ISO 26262 standard, functional validation of the developed Automotive Software Systems (ASSs) is crucial to ensure the safety and reliability aspects. Hardware-in-the-loop (HIL) has been introduced as a reliable, safe and flexible test platform to enable the validation process in real-time. However, the traditional failure analysis process of HIL tests is time-consuming, extremely difficult and requires considerable effort. Therefore, an intelligent solution that can overcome the above challenges is required. Following a data-driven approach, the development of deep learning methods for fault detection and classification has gradually become a hot topic. However, despite the fruitful results, most of the previous studies were conducted for single faults without considering the simultaneous occurrence of multiple faults and ignoring the noisy conditions. In this study, based on multi-label ensemble long short term memory (LSTM) and random forest (RF) techniques, a novel method for simultaneous fault classification under noisy conditions is developed. To improve the robustness of the model against noise, a GRU-based denoising autoencoder (DAE) was implemented. Furthermore, to overcome the challenge of imbalanced data, a random undersampling algorithm was employed. By doing so, the single and simultaneous sensor faults occurring during HIL testing of ASSs can be efficiently and automatically detected and identified. To evaluate the capabilities and robustness of the proposed method, a high-fidelity gasoline engine with a dynamic vehicle system and driving environment was used as a case study. The analysis results demonstrate that the proposed model can achieve a high degree of accuracy under noise with an average detection accuracy of 99.43%. Moreover, compared to the individual methods, the proposed ensemble learning architecture with DAE provides more promising fault identification performance with improved accuracy and robustness. Specifically, the test

关键词： Automotive software systems fault detection and diagnosis deep learning denoising autoencoder LSTM random forest real-time simulation fault injection hardware-in-the-loop (HIL)

来源：评论

学校读者我要写书评

暂无评论

Intelligent denoise laser ultrasonic imaging for inspection of selective laser melting components with rough surface

引用

NDT & E INTERNATIONAL 2022年 125卷 102548-102548页

作者： Xu, Wanli Zhang, Jun Li, Xiaohong Yuan, Shuxian Ma, Guanbing Xue, Zhixiang Jing, Xuechao Cao, Jiancheng Wuhan Univ Sch Power & Mech Engn Wuhan 430072 Hubei Peoples R China CGN Inspect Technol Co Ltd Suzhou 215100 Jiangsu Peoples R China Southeast Univ Sch Mat Sci & Engn Nanjing 211189 Jiangsu Peoples R China

The random microdefects are inevitable during the Selective laser melting (SLM) process due to the principle of discrete-stacking. The rough surface induced strong background noise reduces the probability of detection of traditional laser ultrasonic testing system. In this study, an intelligent denoise laser ultrasonic imaging method was developed to inspect the micro defects on the rough surface of SLM components: (1) a non-contact laser ultrasonic scanning setup was established for data acquisition of multiple ultrasonic signals;(2) a denoising algorithm based on unsupervised machine learning was designed and trained by abundant ultrasonic data to enhance the signal to noise ratio (SNR);(3) a signal matching algorithm based on the cross-correlation and selfnormalized method was established to match the amplitude and arrival time of Rayleigh waves from different scanning points. The performance of developed method was verified using micro hole defects on the rough surface of a SLM part. The results indicated that the established denoising algorithm could increase the average SNR from 27.0 dB to 35.2 dB. All holes with diameter of 50 mu m and 100 mu m can be detected and sized based on the high SNR image without removing the rough surface. The conclusion can be drawn that the proposed intelligent denoise laser ultrasonic imaging method is a very potential way for the online inspection of SLM.

关键词： Laser ultrasonic testing Selective laser melting Surface roughness denoising autoencoder

来源：评论

学校读者我要写书评

暂无评论

DefGAN: Defect Detection GANs With Latent Space Pitting for High-Speed Railway Insulator

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2021年 70卷 1-10页

作者： Zhang, Dongkai Gao, Shibin Yu, Long Kang, Gaoqiang Wei, Xiaoguang Zhan, Dong Southwest Jiaotong Univ Sch Elect Engn Chengdu 610031 Peoples R China

As a key component of the high-speed railway catenary, the insulator plays an important role in supporting the catenary and maintaining the insulation between the catenary and earth. Regular inspection using computer vision technology is an effective way to detect insulator defects and improve the catenary operation safety. However, achieving full automation of insulator defects detection is still a difficult task due to the defective sample scarcity and subtle defect features. To overcome the problem, this article proposes defect detection GANs (DefGANs) that consist of a denoising autoencoder, a discriminator, and a classifier to detect defects, which can generate defective samples from pitted latent representations, thereby improving the reliability of defect detection classifier. The proposed DefGAN can pit the latent representations of normal samples to generate defective samples. Our proposal mainly consists of two stages. In the first stage, we use cascaded deep segmentation networks (CDSNets) to extract the insulator from the catenary images. Then, the insulator area is sampled into small patches. In the second stage, the patches are fed into the proposed DefGAN to detect defects. Moreover, the defect score is determined by the anomaly probability of the classifier and reconstruction error of the denoising autoencoder. The effectiveness of our work is measured across the catenary data sets.

关键词： Anomaly detection classifier denoising autoencoder generator adversarial training insulator latent space

来源：评论

学校读者我要写书评

暂无评论

Signal speech reconstruction and noise removal using convolutional denoising audioencoders with neural deep learning

引用

ANALOG INTEGRATED CIRCUITS AND SIGNAL PROCESSING 2019年第3期100卷 501-512页

作者： Abouzid, Houda Chakkor, Otman Gabriel Reyes, Oscar Ventura, Sebastian Abdelmalek Essaadi Univ Telecommun Dept ENSATE Tetouan Morocco Univ Cordoba Dept Comp Sci & Numer Anal Cordoba Spain

Datasets exist in real life in many formats (audio, music, image,...). In our case, we have them from various sources mixed together. Our mixtures represent noisy audio data that need to be extracted (features), compressed and analysed in order to be presented in a standard way. The resulted data will be used for the Blind Source Separation task. In this paper, we deal with two types of autoencoders: convolutional and denoising. The novelty of our work is to reconstruct the audio signal in the output of the neural network after extracting the meaningful features that present the pure and the powerful information. Simulation results show a great performance, yielding of 87% for the reconstructed signals that will be included in the automated system used for real word applications.

关键词： denoising autoencoder Convolutional autoencoder BSS Keras Deep learning Neural network

来源：评论

学校读者我要写书评

暂无评论

Denoised Residual Trace Analysis for Monitoring Semiconductor Process Faults

引用

IEEE TRANSACTIONS ON SEMICONDUCTOR MANUFACTURING 2019年第3期32卷 293-301页

作者： Jang, Jaeyeon Min, Byung Wook Kim, Chang Ouk Yonsei Univ Dept Ind Engn Seoul 03722 South Korea

The detection of wafer faults in early process steps through monitoring and analyzing multivariate process trace data contribute to wafer yield improvements. Standard classification algorithms have been generally used for fault detection and classification (FDC). However, this approach can cause information loss while extracting statistical features from the trace data and cannot consider class imbalance situations where much fewer faulty wafers are generated than normal wafers. In addition, the approach does not consider normal wafer-to-wafer (W2W) variations and sensor noise inherent in the trace data. These drawbacks significantly degrade FDC performance. This paper proposes a method that builds an FDC model only with trace data of normal wafers in which W2W variations and sensor noise exist. The one-class FDC method detects the occurrence of abnormal trace patterns that cause wafer faults by removing W2W variations and sensor noise from raw traces by using denoising autoencoders, and this method finds the fault-introducing process parameters with the occurrence times. In experiments using the trace data of etch and chemical vapor deposition processes, the proposed method exhibited 1% and 6% higher performance than the best-performing method among comparison methods in terms of the geometric mean of the normal and fault detection accuracies.

关键词： Fault detection and classification denoising autoencoder sensor noise wafer-to-wafer variation multivariate trace data semiconductor manufacturing

来源：评论

学校读者我要写书评

暂无评论

A denoising semi-supervised deep learning model for remaining useful life prediction of turbofan engine degradation

引用

APPLIED INTELLIGENCE 2023年第19期53卷 22682-22699页

作者： Wang, Youming Wang, Yue Xian Univ Posts & Telecommun Sch Automation Xian Key Lab Adv Control & Intelligent Proc Xian 710121 Peoples R China

Remaining useful life (RUL) prediction is significant for reliability analysis and the reduction of maintenance costs for turbofan engine systems. However, most of the existing methods capture temporal or spatial features to predict RUL, which leads to the neglect of deep spatio-temporal correlation in the prediction data. In addition, there are usually noise interference and few labels in feature extraction, which causes difficulty in designing robust RUL prediction models. To address these problems, a denoised semi-supervised model based on fully convolutional denoising autoencoder, convolutional neural network, and long short-term memory network (FCDAE-CNN-LSTM) is proposed to predict RUL, where FCDAE is constructed to denoise the original data by parameter fine-tuning and CNN-LSTM is established based on parallel connection for RUL prognosis. The denoising and generalization capabilities of the RUL prediction model are enhanced by the combination of unsupervised denoising and supervised feature extraction in the case of few labeled training data. The superior features of the method lie in that FCDAE effectively captures globalized and localized features and CNN-LSTM captures multi-layer fused spatial-temporal correlations. The Root Mean Square Error (RMSE) and Score of the proposed method on C-MAPSS dataset are 12.01, 16.62, 11.84, 18.21, and 209.44, 1466.03, 205.07, 2338.93, respectively, which have demonstrated that our method achieves the state-of-the-art performance and outperforms other models.

关键词： Convolutional neural network denoising autoencoder Long short-term memory network Remaining useful life Semi-supervised deep learning

来源：评论

学校读者我要写书评

暂无评论

Bi-channel hybrid GAN attention based anomaly detection system for multi-domain SDN environment

引用

JOURNAL OF INTELLIGENT & FUZZY SYSTEMS 2024年第1期46卷 457-478页

作者： Prabu, Saranya Padmanabhan, Jayashree Anna Univ Dept Comp Technol MIT Campus Chennai Tamil Nadu India

Software-Defined Networking (SDN) is a strategy that leads the network via software by separating its control plane from the underlying forwarding plane. In support of a global digital network, multi-domain SDN architecture emerges as a viable solution. However, the complex and ever-evolving nature of network threats in a multi-domain environment presents a significant security challenge for controllers in detecting abnormalities. Moreover, multi-domain anomaly detection poses a daunting problem due to the need to process vast amounts of data from diverse domains. Deep learning models have gained popularity for extracting high-level feature representations from massive datasets. In this work, a novel deep neural network architecture, supervised learning based LD-BiHGA (Low Dimensional Bi-channel Hybrid GAN Attention) system is designed to learn class-specific features for accurate anomaly detection. Two asymmetric GANs are employed for learning the normal and abnormal network flows separately. Then, to extract more relevant features, a bi-channel attention mechanism is added. This is the first study to introduce an innovative hybrid architecture that merges bi-channel hybrid GANs with attention models for the purpose of anomaly detection in a multi-domain SDN environment that effectively handles real-time unbalanced data. The suggested architecture demonstrates its effectiveness on three benchmark datasets, achieving an average accuracy improvement of 7.225% on balanced datasets and 3.335% on imbalanced datasets compared to previous intrusion detection system (IDS) architectures in the literature.

关键词： Hybrid GAN intrusion detection deep learning attention model dimensionality reduction denoising autoencoder

来源：评论

学校读者我要写书评

暂无评论

Unsupervised environmental operating condition compensation strategies in a guided ultrasonic wave monitoring system: evaluation and comparison

引用

JOURNAL OF CIVIL STRUCTURAL HEALTH MONITORING 2024年第4期14卷 861-884页

作者： Yon, Kong Chen Bakhary, Norhisham Padil, Khairul Hazman Shamsudin, Mohd Fairuz Univ Teknol Malaysia Fac Civil Engn Utm Skudai 81310 Johor Malaysia Univ Teknol Malaysia Fac Mech Engn Utm Skudai 81310 Johor Malaysia Univ Teknol Malaysia Inst Noise & Vibrat City CampusJalan Semarak Kuala Lumpur 54100 Malaysia Univ Teknol Malaysia Inst Smart Infrastruct & Innovat Construct Forens Engn Ctr Johor Baharu 81310 Johor Malaysia VRG Dinamik SDN BHD Shah Alam 40170 Selangor Malaysia

Guided ultrasonic wave (GUW) monitoring systems are gaining much attention in pipeline condition monitoring. However, the effects of environmental and operational conditions (EOCs), especially temperature and random noise, degrade damage detection performance. When EOC effects produce greater amplitudes than the reflected waves from small damage cases, the reflected waves remain unidentified. This paper proposes an unsupervised learning-based denoising autoencoder (DAE) to reduce the effect of EOCs in GUW monitoring systems. A DAE decodes high-dimensional data into low-dimensional features and reconstructs the original data from these low-dimensional features. By providing GUW signals at a reference temperature, this structure forces the DAE to learn the essential features hidden within complex data. The proposed DAE undergoes comparative analysis with other popular unsupervised learning algorithms used for EOC compensation in GUW monitoring systems, such as principal component analysis, independent component analysis and deep autoencoder algorithms. EOC compensation performance is evaluated through receiver operating characteristics (ROC). From the numerical model and an experimental model, the GUW database is obtained. All four algorithms showed good damage detection performance using a numerical model;however, in the experimental model, the proposed DAE showed superiority among other methods.

关键词： Guided ultrasonic wave Unsupervised learning Environmental and operational conditions Independent component analysis Principal component analysis Deep autoencoder denoising autoencoder

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：