检索结果-内蒙古大学图书馆

43rd Chinese Control Conference (CCC)

作者： Zeng, Jianxin Mei, Lin Li, Shuaiyong Li, MaoYang Yang, Zhang Chongqing Univ Posts & Telecommun Minist Educ Key Lab Ind Internet Things & Networked Control Chongqing 400065 Peoples R China Chongqing Special Equipment Inspect & Res Inst Chongqing 401121 Peoples R China

ISBN: (纸本)9798350366907;9789887581581

Ultrasonic guided waves are widely used signals in industry, but their inherent dispersion and multimodal characteristics significantly impact practical applications. Considering the frequency-dependent nature of guided wave signals, this paper proposes a Fourier-based convolutional autoencoder (FCAE) algorithm for dispersion compensation in ultrasonic guided wave signals, thoroughly investigating its practical application in handling ultrasonic guided wave signals. The neural network model combines the time-frequency analysis characteristics of Fourier bases and the nonlinear mapping capability of neural networks to learn the time of flight (TOF) of guided wave modes. Through this approach, the model successfully eliminates phase distortion and energy leakage caused by Lamb wave dispersion in guided wave signals, thereby significantly improving the quality and interpretability of the signals. The proposed Fourier-based neural network model's effectiveness in dispersion compensation is validated through simulation experiments, providing robust support for further applications of ultrasonic guided wave technology.

关键词： Fourier Basis convolutional autoencoder Ultrasonic Guided Wave Dispersion Compensation

来源：评论

学校读者我要写书评

暂无评论

A Feature Extraction Method Based on convolutional autoencoder for Plant Leaves Classification 2nd

A Feature Extraction Method Based on Convolutional Autoencod...

引用

2nd IEEE Colombian Conference on Computational Intelligence (ColCACI)

作者： Paco Ramos, Mery M. Paco Ramos, Vanessa M. Loaiza Fabian, Arnold Osco Mamani, Erbert F. Univ Nacl Jorge Basadre Grohmann Tacna Peru

ISBN: (纸本)9783030362119;9783030362102

In this research, we present an approach based on convolutional autoencoder (CAE) and Support Vector Machine (SVM) for leaves classification of different trees. While previous approaches relied on image processing and manual feature extraction, the proposed approach operates directly on the image pixels, without any preprocessing. Firstly, we use multiple layers of CAE to learn the features of leaf image dataset. Secondly, the extracted features were used to train a linear classifier based on SVM. Experimental results show that the classifiers using these features can improve their predictive value, reaching an accuracy rate of 94.74%.

关键词： Feature extraction Image processing Plant classification convolutional autoencoder Deep learning Computer vision

来源：评论

学校读者我要写书评

暂无评论

A Class-Imbalanced Study with Feature Extraction via PCA and convolutional autoencoder 23

A Class-Imbalanced Study with Feature Extraction via PCA and...

引用

23rd IEEE International Conference on Information Reuse and Integration for Data Science (IEEE IRI)

作者： Salekshahrezaee, Zahra Leevy, Joffrey L. Khoshgoftaar, Taghi M. Florida Atlantic Univ Boca Raton FL 33431 USA

ISBN: (数字)9781665466035

ISBN: (纸本)9781665466035

It is inherently challenging to train a machine learning algorithm on a class-imbalanced dataset. Under conditions of high dimensionality, this training process can become even more difficult due to the large number of features in the dataset. During preprocessing, data sampling is commonly used to address class imbalance and feature extraction is frequently used to reduce the number of dataset features. In this study, we explore the use of these two preprocessing activities before passing on the data to four ensemble classifiers (Random Forest, CatBoost, LightGBM, and XGBoost). With reference to feature extraction, the Principal Component Analysis (PCA) and convolutional autoencoder (CAE) methods are evaluated. With regard to data sampling, the Random Undersampling (RUS) and Synthetic Minority Oversampling Technique (SMOTE) methods are evaluated. Classification performance is measured with the Area Under the Receiver Operating Characteristic Curve (AUC) metric. Our results indicate that the implementation of the RUS method followed by the CAE method leads to the best classification performance.

关键词： feature extraction data sampling class imbalance pca convolutional autoencoder

来源：评论

学校读者我要写书评

暂无评论

Stacked causal convolutional autoencoder based speech compression method 32

Stacked causal convolutional autoencoder based speech compre...

引用

32nd IEEE Signal Processing and Communications Applications Conference (SIU)

作者： Bekiryazici, Tahir Aydemir, Gurkan Gurkan, Hakan Bursa Tekn Univ Elekt Elekt Muhendisligi Bolumu Bursa Turkiye

ISBN: (纸本)9798350388978;9798350388961

This study proposes a speech compression method based on one-dimensional convolutional autoencoder and residual vector quantization. The proposed method offers different compression ratios at low bit rates. Speech quality evaluation metric (PESQ) was used to test the performance of the proposed method. Experimental results show that the proposed method achieves a PESQ value of 1.903 for 2.5 kbps and 2.24 for 5 kbps.

关键词： Speech compression residual vector quantization convolutional autoencoder deep learning

来源：评论

学校读者我要写书评

暂无评论

A convolutional autoencoder Approach for Weakly Supervised Anomaly Video Detection 15th

A Convolutional Autoencoder Approach for Weakly Supervised A...

引用

15th International Conference on Computational Collective Intelligence (ICCCI)

作者： Phan Nguyen Duc Hieu Phan Duy Hung FPT Univ Hanoi Vietnam

ISBN: (纸本)9783031414558;9783031414565

Weakly-supervised video anomaly detection uses video-level labels to avoid annotating all frames or segments in the training video. This problem is typically considered as a multiple instance learning problem, the training process aims to learn how to score both abnormal segments and normal segments, and the score of abnormal segments is higher than the score of normal segments. The features are extracted from videos before the training or testing process. Although many models have been proposed and obtained good results, improving the performance of the problem remains a challenge. This study proposes a convolutional autoencoder based approach to reconstruct features, with an assumption that the reconstructed features contain important information early bound to the objective of normal or abnormal video classification (minimizing the cost function). The work is validated on the ShanghaiTech Campus dataset and has produced results that outperform state-of-the-art methods.

关键词： Anomaly Video Detection convolutional autoencoder Multiple Instance Learning

来源：评论

学校读者我要写书评

暂无评论

Robust Deep 3D convolutional autoencoder for Hyperspectral Unmixing with Hypergraph Learning

引用

Journal of Harbin Institute of Technology(New Series) 2021年第5期28卷 1-8页

作者： Peiyuan Jia Miao Zhang Yi Shen Department of Control Science and Engineering Harbin Institute of TechnologyHarbin 150001China

Hyperspectral unmixing aims to acquire pure spectra of distinct substances(endmembers)and fractional abundances from highly mixed *** this paper,a deep unmixing network framework is designed to deal with the noise *** contains two parts:a three⁃dimensional convolutional autoencoder(denoising 3D CAE)which recovers data from noised input,and a restrictive non⁃negative sparse autoencoder(NNSAE)which incorporates a hypergraph regularizer as well as a l2,1⁃norm sparsity constraint to improve the unmixing *** deep denoising 3D CAE network was constructed for noisy data retrieval,and had strong capacity of extracting the principle and robust local features in spatial and spectral domains efficiently by training with corrupted ***,a part⁃based nonnegative sparse autoencoder with l2,1⁃norm penalty was concatenated,and a hypergraph regularizer was designed elaborately to represent similarity of neighboring pixels in spatial *** experiments were conducted on synthetic and real⁃world data,which both demonstrate the effectiveness and robustness of the proposed network.

关键词： deep learning unsupervised unmixing convolutional autoencoder hypergraph hyperspectral data

来源：评论

学校读者我要写书评

暂无评论

Ensemble of One-Class Classifiers Based on Multi-level Hidden Representations Abstracted from convolutional autoencoder for Anomaly Detection 1

引用

31st International Conference on Artificial Neural Networks (ICANN)

作者： Wang, Xin-Tan Liu, Jian-Wei China Univ Petr Dept Automat Coll Informat Sci & Engn Beijing Peoples R China

ISBN: (数字)9783031159343

ISBN: (纸本)9783031159343;9783031159336

Image anomaly detection has recently emerged a large number of methods, which are widely used in industry, medicine and other fields. In this paper, we propose a novel image anomaly detection method, named Ensemble of One-Class Classifiers based on multi-level hidden representations abstracted from convolutional autoencoder (EOCCA), which is a two-stage method. Specifically, in the training process of the model, first, we train the convolutional autoencoder with the goal of minimizing the reconstruction error, thereby extracting the features from different layers of the encoder. Second, we train multiple one-class support vector machines based on the multi-level hidden representations abstracted from convolutional autoencoder. During inference, we ensemble all trained one-class classifiers for anomaly detection. Experimental results on several image anomaly detection benchmark datasets demonstrate that our proposed method is on par or outperforms current state-of-the-art image anomaly detection methods.

关键词： Image anomaly detection Ensemble learning One-class support vector machine convolutional autoencoder

来源：评论

学校读者我要写书评

暂无评论

LEARNING SENSOR-SPECIFIC FEATURES FOR HYPERSPECTRAL IMAGES VIA 3-DIMENSIONAL convolutional autoencoder 37

LEARNING SENSOR-SPECIFIC FEATURES FOR HYPERSPECTRAL IMAGES V...

引用

IEEE International Geoscience and Remote Sensing Symposium (IGARSS)

作者： Ji, Jingyu Mei, Shaohui Hou, Junhui Li, Xu Du, Qian Northwestern Polytech Univ Sch Elect & Informat Xian 710129 Shaanxi Peoples R China City Univ Hong Kong Dept Comp Sci Kowloon Hong Kong Peoples R China Mississippi State Univ Dept Elect & Comp Engn Mississippi State MS 39762 USA

ISBN: (纸本)9781509049516

Deep learning techniques have brought in revolutionary achievements for feature learning of images. In this paper, a novel structure of 3-Dimensional convolutional autoencoder (3D-CAE) is proposed for hyperspectral spatial-spectral feature learning, in which the spatial context is considered by constructing a 3-Dimensional input using pixels in a spatial neighborhood. All the parameters involved in the 3D-CAE are trained without the need of labeled training samples such that feature learning is conducted in an unsupervised fashion. Such unsupervised spatial-spectral feature extraction is also extended to different images from the same sensor to learn sensor-specific features. As a result, spatial-spectral features of hyperspectral images are extracted for a specific sensor under an unsupervised manner. Experimental results on several benchmark hyperspectral datasets have demonstrated that our proposed 3D-CAE are very effective in extracting sensor-specific spatial-spectral features and outperform several state-of-the-art deep learning neural networks in classification application.

关键词： hyperspectral classification deep learning convolutional autoencoder

来源：评论

学校读者我要写书评

暂无评论

HYPERSPECTRAL IMAGE CLASSIFICATION BASED ON SEMI-SUPERVISED DUAL-BRANCH convolutional autoencoder WITH SELF-ATTENTION

HYPERSPECTRAL IMAGE CLASSIFICATION BASED ON SEMI-SUPERVISED ...

引用

IEEE International Geoscience and Remote Sensing Symposium (IGARSS)

作者： Feng, Jie Ye, Zhanwei Li, Di Liang, Yuping Tang, Xu Zhang, Xiangrong Xidian Univ Minist Educ Key Lab Intelligent Percept & Image Understanding Xian 710071 Peoples R China

ISBN: (纸本)9781728163741

Deep learning method shows its powerful classification performance with sufficient available data. However, the labeled data is limited in hyperspectral images (HSIs). Semi-supervised algorithms have unique advantages on dealing with this problem. Therefore, a semi-supervised convolutional neural network is proposed in this paper. It consists of two branches, which use limited labeled samples and a large number of unlabeled samples, respectively. The first branch includes an encoder-decoder model to extract contextual information of unlabeled samples. The other one uses the similar construction except extra classification layers to extract discriminative features of labeled samples. In order to fuse contextual and discriminative information, we cascade the features of low-level layers from different branches. Furthermore, self-attention is added to the first branch, which focuses more on the global information for classification. The experiment results show that the proposed model provides a competitive result compared with state-of-the-art methods.

关键词： hyperspectral images classification semi-supervised learning convolutional autoencoder self-attention mechanism

来源：评论

学校读者我要写书评

暂无评论

UNSUPERVISED IMAGE SEGMENTATION USING convolutional autoencoder WITH TOTAL VARIATION REGULARIZATION AS PREPROCESSING

UNSUPERVISED IMAGE SEGMENTATION USING CONVOLUTIONAL AUTOENCO...

引用

IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Wang, Chunlai Yang, Bin Liao, Yiwen Univ Stuttgart Inst Signal Proc & Syst Theory Stuttgart Germany

ISBN: (纸本)9781509041176

Conventional unsupervised image segmentation methods use color and geometric information and apply clustering algorithms over pixels. They preserve object boundaries well but often suffer from over-segmentation due to noise and artifacts in the images. In this paper, we contribute on a preprocessing step for image smoothing, which alleviates the burden of conventional unsupervised image segmentation and enhance their performance. Our approach relies on a convolutional autoencoder (CAE) with the total variation loss (TVL) for unsupervised learning. We show that, after our CAE-TVL preprocessing step, the over-segmentation effect is significantly reduced using the same unsupervised image segmentation methods. We evaluate our approach using the BSDS500 image segmentation benchmark dataset and show the performance enhancement introduced by our approach in terms of both increased segmentation accuracy and reduced computation time. We examine the robustness of the trained CAE and show that it is directly applicable to other natural scene images.

关键词： Image segmentation unsupervised convolutional autoencoder total variation loss

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：