检索结果-内蒙古大学图书馆

An Integral Projection-Based Semantic autoencoder for Zero-Shot Learning

IEEE ACCESS 2023年 11卷 85351-85360页

作者： Heyden, William Ullah, Habib Siddiqui, Muhammad Salman Al-Machot, Fadi Norwegian Univ Life Sci NMBU Fac Sci & Technol REALTEK N-1430 As Norway

Zero-shot Learning (ZSL) classification categorizes or predicts classes (labels) that are not included in the training set (unseen classes). Recent works proposed different semantic autoencoder (SAE) models where the encoder embeds a visual feature vector space into the semantic space and the decoder reconstructs the original visual feature space. The objective is to learn the embedding by leveraging a source data distribution, which can be applied effectively to a different but related target data distribution. Such embedding-based methods are prone to domain shift problems and are vulnerable to biases. We propose an integral projection-based semantic autoencoder (IP-SAE) where an encoder projects a visual feature space concatenated with the semantic space into a latent representation space. We force the decoder to reconstruct the visual-semantic data space. Due to this constraint, the visual-semantic projection function preserves the discriminatory data included inside the original visual feature space. The enriched projection forces a more precise reconstitution of the visual feature space invariant to the domain manifold. Consequently, the learned projection function is less domain-specific and alleviates the domain shift problem. Our proposed IP-SAE model consolidates a symmetric transformation function for embedding and projection, and thus, it provides transparency for interpreting generative applications in ZSL. Therefore, in addition to outperforming state-of-the-art methods considering four benchmark datasets, our analytical approach allows us to investigate distinct characteristics of generative-based methods in the unique context of zero-shot inference.

关键词： autoencoder generative modelling generative regularisation latent space linear transfor-mation semantic embedding visual projection zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

Residual-recursive autoencoder for accelerated evolution in savonius wind turbines optimization

引用

NEUROCOMPUTING 2022年第0期500卷 909-920页

作者： Zhou, Qianwei Li, Baoqing Tao, Peng Xu, Zhang Zhou, Chen Wu, Yanzhuang Hu, Haigen Zhejiang Univ Technol Coll Comp Sci & Technol Hangzhou Peoples R China Key Lab Visual Media Intelligent Proc Technol Zhej Hangzhou Peoples R China Zhejiang Univ Technol Coll Mech Engn Hangzhou Peoples R China Chinese Acad Sci Shanghai Inst Microsyst & Informat Technol Beijing Peoples R China

Recent studies verified that a genetic algorithm can discover efficient and innovative wind turbines by using image encoding and decoding techniques. To accelerate the optimization, in this work, ResidualRecursion autoencoder (RRAE) is proposed to extract low-dimensional latent codes from rotors' crosssection images while maintaining reconstruction accuracy as high as possible. As a kind of neural network framework, the advantages of using RRAE are threefold: 1) RRAE can wrap over different kinds of autoencoders and improve their performance;2) RRAE is compatible with different kinds of loss functions and works well with very low-dimensional latent codes;3) RRAE is easy to use and efficient in decoding latent codes which is important to the rapid convergence of the genetic algorithm. The experiment results has shown that the reconstruction loss has decreased by 30.56% on a recursive autoencoder, 11.40% to 29.34% on different feedforward autoencoders. Two RRAE-accelerated optimizations have been carried out in this work. One has used only 14% of the calculation required by the baseline method without any deterioration in rotor performance. The other one has used 52.33% and increased the rotor performance by 7.59%.

关键词： Recursive neural network autoencoder Deep learning Geometry optimization Savonius rotor

来源：评论

学校读者我要写书评

暂无评论

An autoencoder-Based I/Q Channel Interaction Enhancement Method for Automatic Modulation Recognition

引用

IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY 2023年第7期72卷 9620-9625页

作者： Zhang, Fuxin Luo, Chunbo Xu, Jialang Luo, Yang Univ Elect Sci & Technol China Sch Informat & Commun Engn Chengdu 611731 Peoples R China

This article proposes an autoencoder-based method to enhance the information interaction between in-phase/quadrature (I/Q) channels of the input data for automatic modulation recognition (AMR). The proposed method utilizes an autoencoder built by fully-connected layers to correlate the features of I/Q data and obtain the interaction feature from the intermediate layer, which is concatenated together with the original I/Q data as model inputs. To accommodate the new data dimensions, a modification scheme for the existing representative deep learning based AMR (DL-AMR) models is presented. Experimental results show that our method can improve the recognition accuracy of the state-of-the-art baseline models, and has a smaller time overhead compared with complex-valued neural networks.

关键词： Automatic modulation recognition in-phase quadrature signal deep learning autoencoder

来源：评论

学校读者我要写书评

暂无评论

Comparing autoencoder Variants for Real-Time Denoising of Hyperspectral X-Ray

引用

IEEE SENSORS JOURNAL 2022年第18期22卷 17997-18007页

作者： Bonettini, Nicolo Gonano, Carlo Andrea Bestagini, Paolo Marcon, Marco Garavelli, Bruno Tubaro, Stefano Politecn Milan Comp Sci & Engn Milan Italy Politecn Milan Elect Engn Milan Italy Politecn Milan Informat Technol Milan Italy Informat & Commun Technol ICT Trieste Italy

Hyperspectral X ray analysis is used in many industrial pipelines, from quality control to detection of low-density contaminants in food. Unfortunately, the signal acquired by X-ray sensors is often affected by a great amount of noise. This hinders the performance of most of the applications building on top of these acquisitions (e.g., detection of food contaminants). Therefore, a good denoising pipeline is necessary. This article proposes a comparison between three different autoencoder variants: the Variational autoencoder, the Augmented autoencoder, and a plain vanilla autoencoder. All the networks are trained in an unsupervised fashion to denoise a given noisy spectrum. Focusing on the specific application of recognizing possible food contaminants, we force the latent space of the networks to have just two parameters, as suggested by the physical law of Lambert- Beer. We validate our experiments on a synthetic dataset composed of roughly 15 million spectra. Results suggest that the Augmented autoencoder is the best network configuration for this task, showing excellent performance without suffering from the nondeterministic behavior of the Variational autoencoder.

关键词： autoencoder denoising hyperspectral X-ray

来源：评论

学校读者我要写书评

暂无评论

An end-to-end multi-scale network based on autoencoder for infrared and visible image fusion

引用

MULTIMEDIA TOOLS AND APPLICATIONS 2023年第13期82卷 20139-20156页

作者： Liu, Hongzhe Yan, Hua Sichuan Univ Coll Elect & Informat Engn Chengdu 610065 Sichuan Peoples R China

Infrared and visible image fusion aims to obtain a more informative fusion image by merging the infrared and visible images. However, the existing methods have some shortcomings, such as detail information loss, unclear boundaries, and not being end-to-end. In this paper, we propose an end-to-end network architecture for infrared and visible image fusion task. Our network contains three essential parts: encoders, residual fusion module, and decoder. First, we input infrared and visible images to two encoders to extract shallow features, respectively. Subsequently, the two sets of features are concatenated and fed to the residual fusion module to extract multi-scale features and fuse them adequately. Finally, the fused image is obtained by the decoder. We conduct objective and subjective experiments on two public datasets. The comparison results with the state-of-art methods prove that the fusion results of the proposed method have better objective metrics and contain more detail information and more explicit boundary.

关键词： autoencoder Image fusion Infrared image Visible image

来源：评论

学校读者我要写书评

暂无评论

Feature Learning With a Divergence-Encouraging autoencoder for Imbalanced Data Classification

引用

IEEE ACCESS 2018年 6卷 70197-70211页

作者： Luo, Ruisen Feng, Qian Wang, Chen Yang, Xiaomei Tu, Haiyan Yu, Qin Fei, Shaomin Gong, Xiaofeng Sichuan Univ Coll Elect Engn & Informat Technol Chengdu 610064 Sichuan Peoples R China UCL Dept Comp Sci London WC1E 6BT England Chengdu Univ Informat Technol Expt Ctr Elect Chengdu 610059 Sichuan Peoples R China DaGongBoChuang Corp Chengdu 610005 Sichuan Peoples R China

Imbalanced data exists commonly in machine learning classification applications. Popular classification algorithms are based on the assumption that data in different classes are roughly equally distributed;however, extremely skewed data, with instances from one class taking up most of the dataset, is not exceptional in practice. Thus, performance of algorithm often degrades significantly when encountering skewed data. Mitigating the problem caused by imbalanced data has been an open challenge for years, and previous researches mostly have proposed solutions from the perspectives of data re-sampling and algorithm improvement. In this paper, focusing on two-class imbalanced data, we have proposed a novel divergence-encouraging autoencoder (DEA) to explicitly learn features from both of the two classes and have designed an imbalanced data classification algorithm based on the proposed autoencoder. By encouraging maximization of divergence loss between different classes in the bottleneck layer, the proposed DEA can learn features for both majority and minority classes simultaneously. The training procedure of the proposed autoencoder is to alternately optimize reconstruction and divergence losses. After obtaining the features, we directly compute the cosine distances between the training and testing features and compare the median of distances between classes to perform classification. Experimental results illustrate that our algorithm outperform ordinary and loss-sensitive CNN models both in terms of performance evaluation metrics and convergence properties. To the best of our knowledge, this is the first paper proposed to solve the imbalanced data classification problem from the perspective of explicitly learning representations of different classes simultaneously. In addition, designing of the proposed DEA is also an innovative work, which could improve the performance of imbalanced data classification without data re-sampling and benefit future researches in the fie

关键词： Imbalanced data classification autoencoder divergence loss convergence analysis alternating training paradigm

来源：评论

学校读者我要写书评

暂无评论

Effectiveness of autoencoder for lake area extraction from high-resolution RGB imagery: an experimental study

引用

ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH 2021年第24期28卷 31084-31096页

作者： Tercan, Emre Atasever, Umit Haluk Gen Directorate Highways Dept Survey Project & Environm 13th Reg TR-07090 Antalya Turkey Erciyes Univ Fac Engn Dept Geomat Engn TR-38039 Kayseri Turkey

The surface areas of lakes alter constantly due to many factors such as climate change, land use policies, and human interventions, and their surface areas tend to decrease. It is necessary for obtain baseline datasets such as surface areas and boundaries of water bodies with high accuracy, effectively, economically, and practically by using satellite images in terms of management and planning of lakes. Extracting surface areas of water bodies using image classification algorithms and high-resolution RGB satellite images and evaluating the effectiveness of different image classification algorithms have become an important research domain. In this experimental study, eight different machine learning-based classification approaches, namely, k-nearest neighborhood (kNN), subspaced kNN, support vector machines (SVMs), random forest (RF), bagged tree (BT), Naive Bayes (NB), and linear discriminant (LD), have been utilized to extract the surface areas of lakes. Lastly, autoencoder (AE) classification algorithm was applied, and the effectiveness of all those algorithms was compared. Experimental studies were carried out on three different lakes (Hazar Lake, Salda Lake, Manyas Lake) using high-resolution Turkish RASAT RGB satellite images. The results indicated that AE algorithm obtained the highest accuracy values in both quantitative and qualitative analyses. Another important aspect of this study is that Structural Similarity Index (SSIM) and Universal Image Quality Index (UIQI) metrics that can evaluate close to human perception are used for comparison. With this application, it has been shown that overall accuracy calculated from test data may be inadequate in some cases by using SSIM, UIQI, mean squared error (MSE), peak signal to noise ratio (PSNR), and Cohen's KAPPA metrics. In the last application, the robustness of AE was examined with boxplots.

关键词： Machine learning autoencoder Remote sensing Image classification Water body extraction RASAT Environmental monitoring Lake management

来源：评论

学校读者我要写书评

暂无评论

Densely connected convolutional network block based autoencoder for panorama map compression

引用

SIGNAL PROCESSING-IMAGE COMMUNICATION 2020年 80卷 115678-000页

作者： Wang, Shengwei Wang, Hongkui Xiang, Sen Yu, Li Huazhong Univ Sci & Technol Sch Electron Inform & Commun Wuhan 430074 Hubei Peoples R China Wuhan Univ Sci & Tech Sch Inform Sci & Engn Wuhan 430081 Hubei Peoples R China

As a novel virtual reality (VR) format, panorama maps are attracting increasing attention, while the compression of panorama images is still a concern. In this paper, a densely connected convolutional network block (dense block) based autoencoder is proposed to compress panorama maps. In the proposed autoencoder, dense blocks are specially designed to reuse feature maps and reduce redundancy of features. Meanwhile, a loss function, which imports a position-dependent weight item for each pixel, is proposed to train and adjust network parameters, in order to make the autoencoder fit to properties of panorama maps. Based on the proposed autoencoder and the weighted loss function, a greedy block-wise training scheme is also designed to avoid gradient vanishing problem and speed up training. During training process, the autoencoder is divided into several sub-nets. After each sub-net is trained separately, the whole network is fine-tuned to achieve the best performance. Experimental results demonstrate that the proposed autoencoder, compared with JPEG, saves up to 79.69 % bit rates, and obtains 7.27dB gain in BD-WS-PSNR or 0.0789 gain in BD-WS-SSIM. The proposed autoencoder also outperforms JPEG 2000, HEVC and VVC in both BD-WS-PSNR and BD-WS-SSIM. Meanwhile, subjective results show that the proposed autoencoder can recover details of panorama images, and reconstruct maps with high visual quality.

关键词： autoencoder Dense block Neural network Panorama map

来源：评论

学校读者我要写书评

暂无评论

BSSNet: A Real-Time Semantic Segmentation Network for Road Scenes Inspired From autoencoder

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 2024年第5期34卷 3424-3438页

作者： Shi, Xiaoqiang Yin, Zhenyu Han, Guangjie Liu, Wenzhuo Qin, Li Bi, Yuanguo Li, Shurui Univ Chinese Acad Sci Sch Comp Sci & Technol Beijing 100049 Peoples R China Chinese Acad Sci Shenyang Inst Comp Technol Shenyang 110168 Peoples R China Liaoning Key Lab Domest Ind Control Platform Techn Shenyang 110168 Peoples R China Hohai Univ Dept Internet Things Engn Changzhou 213022 Peoples R China China Univ Min & Technol Beijing Sch Artificial Intelligence Beijing 100083 Peoples R China Northeastern Univ Sch Comp Sci & Engn Shenyang 110167 Peoples R China Shenyang Aerosp Univ Sch Comp Sci Shenyang 110136 Peoples R China

Although semantic segmentation methods have made remarkable progress so far, their long inference process limits their use in practical applications. Recently, some two-branch and three-branch real-time segmentation networks have been proposed to improve segmentation accuracy by adding branches to extract spatial or border information. For the design of extracting spatial information branches, preserving high-resolution features or adding segmentation loss to guide spatial branches are commonly used methods to extract spatial information. However, these approaches are not the most efficient. To solve the problem, we design the spatial information extraction branch as an autoencoder structure, which allows us to extract the spatial structure and features of the image during the encoding and decoding process of the autoencoder. Border, semantic and spatial information are all helpful for segmentation tasks, and efficiently fusing these three kinds of information can obtain better feature representation compared to the fusion of two types of information in the dual-branch network. However, existing three-branch networks have yet to explore this aspect deeply. Therefore, this paper designs a new three-branch network based on this starting point. In addition, we also propose a feature fusion module called the Unified Multi-Feature Fusion module (UMF), which can fuse multiple features efficiently. Our method achieves a state-of-the-art trade-off between inference speed and accuracy on the Cityscapes, CamVid, and NightCity datasets. Specifically, BSSNet-T achieves 78.8% mIoU at 115.8 FPS on the Cityscapes dataset, 79.5% mIoU at 170.8 FPS on the CamVid dataset, and 52.6% mIoU at 172.3 FPS on the NightCity dataset. Code is available at https://***/SXQ-STUDY/BSSNet.

关键词： Real-time semantic segmentation convolution neural networks autoencoder feature fusion

来源：评论

学校读者我要写书评

暂无评论

A digital twin framework for anomaly detection in industrial robot system based on multiple physics-informed hybrid convolutional autoencoder

引用

JOURNAL OF MANUFACTURING SYSTEMS 2024年 77卷 798-809页

作者： Wang, Shijie Tao, Jianfeng Jiang, Qincheng Chen, Wei Qin, Chengjin Liu, Chengliang Shang Hai Jiao Tong Univ State Key Lab Mech Syst & Vibrat Shanghai 201100 Peoples R China Shanghai STEP Robot Co Ltd Shanghai 201800 Peoples R China

The robot number in industry is growing up rapidly. Building anomaly detection system for them can improve the security of these expensive devices. The article implements an anomaly detection framework based on digital twin, which are built by a hybrid convolutional autoencoder. The framework shares those neural network weight files as digital assets, users can use them to estimate the possible output from real input. It approximates the dynamic relationship between motion, current, temperature and vibration with hybrid convolution. Considering the limited generalization performance of direct data-driven methods in practical physical systems, this article introduces physical information methods to improve the constraint function of neural network. The influence of multiple physical fields on current is established by a unified neural network. Terminals detect anomaly with KL divergence between really current and estimated current. The article collects operational data from real robots and verifies it, and the experiment shows that the RMSE for current estimation is below 1.5 %, the F1-score in anomaly detection is over 98.23 %, false positive is below 1 %, false negative is below 1.7 %. The relevant technologies are gradually being promoted and applied in enterprises.

关键词： Anomaly detection Digital twin Industrial robot Physics-Informed 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：