检索结果-南通市图书馆

Hyperspectral image super-resolution Meets Deep Learning:A Survey and Perspective

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

IEEE/CAA Journal of Automatica Sinica 2023年第8期10卷 1668-1691页

作者： Xinya Wang Qian Hu Yingsong Cheng Jiayi Ma the Electronic Information School Wuhan UniversityWuhan 430072China IEEE

Hyperspectral image super-resolution,which refers to reconstructing the high-resolution hyperspectral image from the input low-resolution observation,aims to improve the spatial resolution of the hyperspectral image,which is beneficial for subsequent *** development of deep learning has promoted significant progress in hyperspectral image super-resolution,and the powerful expression capabilities of deep neural networks make the predicted results more ***,several latest deep learning technologies have made the hyperspectral image super-resolution method ***,a comprehensive review and analysis of the latest deep learning methods from the hyperspectral image super-resolution perspective is *** this end,in this survey,we first introduce the concept of hyperspectral image super-resolution and classify the methods from the perspectives with or without auxiliary ***,we review the learning-based methods in three categories,including single hyperspectral image super-resolution,panchromatic-based hyperspectral image super-resolution,and multispectral-based hyperspectral image ***,we summarize the commonly used hyperspectral dataset,and the evaluations for some representative methods in three categories are performed qualitatively and ***,we briefly introduce several typical applications of hyperspectral image super-resolution,including ground object classification,urban change detection,and ecosystem ***,we provide the conclusion and challenges in existing learning-based methods,looking forward to potential future research directions.

关键词： Deep learning hyperspectral image image fusion image super-resolution survey

In-situ laser-induced surface damage inspection based on image super-resolution and adaptive segmentation method

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Chinese Optics Letters 2022年第7期20卷 7-12页

作者： Fengdong Chen Jingyang Sun Qian Wang Hongbo Zhu Fa Zeng Yueyue Han Cheng Lu Guodong Liu Instrument Science and Technology Harbin Institute of TechnologyHarbin 150001China Research Center of Laser Fusion China Academy of Engineering PhysicsMianyang 621900China

In-situ laser-induced surface damage inspection plays a key role in protecting the large aperture optics in an inertial confinement fusion(ICF)high-power laser *** order to improve the initial damage detection capabilities,an in-situ inspection method based on image super-resolution and adaptive segmentation method is *** transfer learning and integration of various attention mechanisms,the super-resolution reconstruction of darkfield images with less texture information is effectively realized,and,on the basis of image super-resolution,an adaptive image segmentation method is designed,which effectively adapts to the damage detection problems under conditions of uneven illumination and weak *** online experiment was carried out by using edge illumination and the telescope optical imaging system,and the validity of the method was proved by the experimental results.

关键词： laser optic damage image super-resolution image segmentation

Fully 1×1 Convolutional Network for Lightweight image super-resolution

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Machine Intelligence Research 2024年第6期21卷 1062-1076页

作者： Gang Wu Junjun Jiang Kui Jiang Xianming Liu Faculty of Computing Harbin Institute of TechnologyHarbin150001China

Deep convolutional neural networks,particularly large models with large kernels(3x3 or more),have achieved significant progress in single image super-resolution(SISR)***,the heavy computational footprint of such models prevents their de-ployment in real-time,resource-constrained ***,1×1 convolutions have substantial computational efficiency,but struggle with aggregating local spatial representations,which is an essential capability for SISR *** response to this dichotomy,we propose to harmonize the merits of both 3x3 and 1×1 kernels,and exploit their great potential for lightweight SISR ***-ally,we propose a simple yet effective fully 1×1 convolutional network,named shift-Conv-based network(SCNet).By incorporating a parameter-free spatial-shift operation,the fully 1×1 convolutional network is equipped with a powerful representation capability and impressive computational *** experiments demonstrate that SCNets,despite their fully 1×1 convolutional structure,consistently match or even surpass the performance of existing lightweight SR models that employ regular *** code and pretrained models can be found at .

关键词： Lightweight network image super-resolution convolutional neural network transformer image restoration

Fractional-order difference curvature-driven fractional anisotropic diffusion equation for image super-resolution

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

International Journal of Modeling, Simulation, and Scientific Computing 2019年第1期10卷 177-189页

作者： Xuehui Yin Shunli Chen Liping Wang Shangbo Zhou School of Software and Engineering Chongqing University of Posts and Telecommunications Chongqing 400065 P.R.China Key Laboratory of Dependable Service Computing in Cyber Physical Society Ministry of Education Chongqing University Chongqing 400030P.R.China College of Computer Science Chongqing University Chongqing 400030P.R.China

image super-resolution methods-based existing edge indicating operators—namely Gauss curvature,mean curvature and gradient-cannot effectively identify the edges,ramps and flat regions and suffer from the loss of fine *** address these issues,this paper presents a fractional anisotropic diffusion equation based on a new edge indicator,named fractional-order difference curvature,which can characterize the intensity variations in *** introduce the frequency-domain definition for fractional-order derivative by the Fourier transform,which is easy to implement *** new edge indicator is better than the existing edge indicating operators in distinguishing between ramps and edges and can better handle the fine *** results for natural images validate that the proposed method can yield a visually pleasing result and better values of MSSIM and PSNR.

关键词： image super-resolution fractional differentiation difference curvature anisotropic diffusion

Enhanced Full-resolution Residual Network for image super-resolution

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Enhanced Full-Resolution Residual Network for Image Super-Re...

第40届中国控制会议

作者： Jiaoyue Li Lifei Zhao Qianqian Shao Weifeng Liu Kai Zhang Bao-Di Liu College of Oceanography and Space Informatics China University of Petroleum (East China) College of Control Science and Engineering China University of Petroleum (East China) College of Petroleum Engineering China University of Petroleum (East China)

Convolutional neural network（CNN） has played a critical role in promoting image super-resolution（SR） performance,and researchers have proposed various models in recent years. Although these models can improve the resolution of the image,the sharpness is not ideal. In this paper, We modify a semantic segmentation network: Full-resolution Residual Network（FRRN）and propose an Enhanced Full-resolution Residual Network for image super-resolution（EFRN） model to improve the image clarity. Our network mainly applies a long skip connection to the direct fusion of low-level and high-level features. Besides, we also introduce the attention mechanism into the model to enhance the network’s repair capability. Moreover, experimental results confirm that EFRN has achieved better accuracy and visual effects than state-of-the-art methods.

关键词： Enhanced Full-resolution Residual Block Spatial Attention Channel Attention image super-resolution

Residual Net Use on FSRCNN for image super-resolution

在线全文

cnki会议

学校读者我要写书评

暂无评论

Residual Net Use on FSRCNN for Image Super-Resolution

第40届中国控制会议

作者： Jinlu Zhang Mingliang Liu Xiaohang Wang Chengcheng Cao Department of Automation Heilongjiang University Key Laboratory of Information Fusion Estimation and Detection Heilongjiang Province

In this paper,we present a new method based on Accelerating the super-resolution Convolutional Neural Network（FSRCNN） and residual network for the aim of image *** our works,we fuse the network structure of FSRCNN and skip connections as the first step of our method,this part can transfer the original features to the training results effectively,in addition,we also change the deconvolution layer to a sub-pixel convolution *** experimental results show that our method is easy to converge in training works and the image reconstruction result also has better *** proposed method has better effects corresponding to the reconstruct images on the Set5,Set14 and General-100 data sets,its average PSNR can increase about 0.38dB.

关键词： image super-resolution Convolutional Neural Network Residual Network

Single image super-resolution via the Implementation of the Hardware-Friendly Sparse Coding

在线全文

cnki会议

学校读者我要写书评

暂无评论

Single Image Super-Resolution via the Implementation of the ...

第37届中国控制会议

作者： Zhekang Dong Chun Sing Lai Zhao Xu Donglian Qi College of Electrical Engineering Zhejiang University Department of Electrical Engineering The Hong Kong Polytechnic University Department of Engineering Science University of Oxford

Sparse representation is one of the effective approaches to perform the feature extraction on the high-dimensional signals and has been widely used in the field of image processing. In this paper, a novel single image super-resolution（SR） reconstruction algorithm based on sparse coding is presented. Specifically, due to the fact that sparse coding has the potential to greatly facilitate the biological-neural system and to deal with a large amount of complex data, while consuming a finite amount of energy, a memristor crossbar array synthesized by a threshold-type memristor model is constructed for the implementation of a hardware-friendly sparse coding method, i.e., the Locally Competitive Algorithm（LCA）. Meanwhile, the relevant online dictionary learning method is provided, along with the description of weight programming strategy. Furthermore, the specific description of the image SR via the LCA sparse coding is provided. For the verification purpose, a series of experiments are carried out to illustrate the validity and effectiveness of the entire scheme.

关键词： image super-resolution sparse coding memristor crossbar array dictionary learning

Optimization of Single image super-resolution Reconstruction Algorithm Based on Residual Dense Network

在线全文

cnki会议

学校读者我要写书评

暂无评论

Optimization of Single Image Super-Resolution Reconstruction...

第32届中国控制与决策会议

作者： Rui Zhang Siqi Wang Zi’ang Wu Shenyang Jianzhu University Information ＆ Control Engineering Faculty

Aiming at the problem of low performance of the Single image super-resolution（SISR） method,a method based on residual dense network is *** shallow convolution features are input to the residual dense block to obtain local and global features,and the reconstruction is performed on the images to obtain a clear high-resolution *** verify the effectiveness of the method,quantitative experiments were performed on three public datasets including Set5,Set14 and *** results show that this method can better recover high-resolution images.

关键词： image super-resolution Deep learning Low resolution and High resolution

A Fast Dictionary Training Algorithm for Single image super-resolution

在线全文

cnki会议

学校读者我要写书评

暂无评论

A Fast Dictionary Training Algorithm for Single Image Super-...

2013年中国智能自动化学术会议

作者： Yiliang Lv Jiwei Liu School of Automation and Electrical Engineering University of Science and Technology Beijing

Generally the dictionary for single image super-resolution is trained by iterations of MP algorithm and K-SVD *** the dictionary,low resolution images can be restored to high resolution images with high *** the training process always takes a lot of *** in this paper we use SVD to analyze the space relationship between the high and low resolution samples,and present a cluster based algorithm for dictionary *** with the K-SVD based algorithm,the proposed algorithm trains the dictionary with a much higher speed, and restores the images with similar visual quality.

关键词： image super-resolution Cluster SVD Spares representation