检索结果-内蒙古大学图书馆

FAN: Frequency aggregation network for real image super-resolution

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Pang, Yingxue Li, Xin Jin, Xin Wu, Yaojun Liu, Jianzhao Liu, Sen Chen, Zhibo CAS Key Laboratory of Technology in Geo-spatial Information Processing and Application System University of Science and Technology of China Hefei230027 China

Single image super-resolution (SISR) aims to recover the high-resolution (HR) image from its low-resolution (LR) input image. With the development of deep learning, SISR has achieved great progress. However, It is still a challenge to restore the real-world LR image with complicated authentic degradations. Therefore, we propose FAN, a frequency aggregation network, to address the real-world image super-resolution problem. Specifically, we extract different frequencies of the LR image and pass them to a channel attention-grouped residual dense network (CA-GRDB) individually to output corresponding feature maps. And then aggregating these residual dense feature maps adaptively to recover the HR image with enhanced details and textures. We conduct extensive experiments quantitatively and qualitatively to verify that our FAN performs well on the real image super-resolution task of AIM 2020 challenge. According to the released final results, our team SR-IM achieves the fourth place on the X4 track with PSNR of 31.1735 and SSIM of 0.8728. Copyright © 2020, The Authors. All rights reserved.

关键词： Textures

learning disentangled feature representation for Hybrid-distorted image restoration

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Li, Xin Jin, Xin Lin, Jianxin Yu, Tao Liu, Sen Wu, Yaojun Zhou, Wei Chen, Zhibo CAS Key Laboratory of Technology in Geo-spatial Information Processing and Application System University of Science and Technology of China Hefei230027 China

Hybrid-distorted image restoration (HD-IR) is dedicated to restore real distorted image that is degraded by multiple distortions. Existing HD-IR approaches usually ignore the inherent interference among hybrid distortions which compromises the restoration performance. To decompose such interference, we introduce the concept of Disentangled Feature Learning to achieve the feature-level divide-and-conquer of hybrid distortions. Specifically, we propose the feature disentanglement module (FDM) to distribute feature representations of different distortions into different channels by revising gain-control-based normalization. We also propose a feature aggregation module (FAM) with channel-wise attention to adaptively filter out the distortion representations and aggregate useful content information from different channels for the construction of raw image. The effectiveness of the proposed scheme is verified by visualizing the correlation matrix of features and channel responses of different distortions. Extensive experimental results also prove superior performance of our approach compared with the latest HD-IR schemes. Copyright © 2020, The Authors. All rights reserved.

关键词： Image reconstruction

Deep local and global spatiotemporal feature aggregation for blind video quality assessment

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Zhou, Wei Chen, Zhibo CAS Key Laboratory of Technology in Geo-Spatial Information Processing and Application System University of Science and Technology of China Hefei230027 China

In recent years, deep learning has achieved promising success for multimedia quality assessment, especially for image quality assessment (IQA). However, since there exist more complex temporal characteristics in videos, very little work has been done on video quality assessment (VQA) by exploiting powerful deep convolutional neural networks (DCNNs). In this paper, we propose an efficient VQA method named Deep SpatioTemporal video Quality assessor (DeepSTQ) to predict the perceptual quality of various distorted videos in a no-reference manner. In the proposed DeepSTQ, we first extract local and global spatiotemporal features by pre-trained deep learning models without fine-tuning or training from scratch. The composited features consider distorted video frames as well as frame difference maps from both global and local views. Then, the feature aggregation is conducted by the regression model to predict the perceptual video quality. Finally, experimental results demonstrate that our proposed DeepSTQ outperforms state-of-the-art quality assessment algorithms. Copyright © 2020, The Authors. All rights reserved.

关键词： Deep neural networks

A decomposed dual-cross generative adversarial network for image rain removal 29

学校读者我要写书评

暂无评论

A decomposed dual-cross generative adversarial network for i...

29th British Machine Vision Conference, BMVC 2018

作者： Jin, Xin Chen, Zhibo Lin, Jianxin Chen, Jiale Zhou, Wei Shan, Chaowei CAS Key Laboratory of Technology in Geo-spatial Information Processing and Application System University of Science and Technology of China Hefei230027 China

Rain removal is important for many computer vision applications, such as surveillance, autonomous car, etc. Traditionally, rain removal is regarded as a signal removal problem which usually causes over-smoothing by removing texture details in non-rain background regions. This paper considers the issue of rain removal from a completely different perspective, to treat rain removal as a signal decomposition problem. Specifically, we decompose the rain image into two components, namely non-rain background image and rain streaks image. Then, we introduce an adversarial training mechanism to synthesize non-rain background image and rain streaks image in a Dual-Cross manner, which makes the two adversarial branches interact with each other, archiving a win-win result ultimately. The proposed Decomposed Dual-Cross Generative Adversarial Network (DDC-GAN) shows significantly performance improvement compared with state-of-the-art methods on both synthetic and real-world images in terms of qualitative and quantitative measures (over 3dB gains in PSNR). © 2018. The copyright of this document resides with its authors.

关键词： Rain

Image super-resolution quality assessment: Structural fidelity versus statistical naturalness

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Zhou, Wei Wang, Zhou Chen, Zhibo Dept. of Electrical and Computer Engineering University of Waterloo WaterlooONN2L3G1 Canada CAS Key Laboratory of Technology in Geo-spatial Information Processing and Application System University of Science and Technology of China Hefei230027 China

Single image super-resolution (SISR) algorithms reconstruct high-resolution (HR) images with their low-resolution (LR) counterparts. It is desirable to develop image quality assessment (IQA) methods that can not only evaluate and compare SISR algorithms, but also guide their future development. In this paper, we assess the quality of SISR generated images in a two-dimensional (2D) space of structural fidelity versus statistical naturalness. This allows us to observe the behaviors of different SISR algorithms as a tradeoff in the 2D space. Specifically, SISR methods are traditionally designed to achieve high structural fidelity but often sacrifice statistical naturalness, while recent generative adversarial network (GAN) based algorithms tend to create more natural-looking results but lose significantly on structural fidelity. Furthermore, such a 2D evaluation can be easily fused to a scalar quality prediction. Interestingly, we find that a simple linear combination of a straightforward local structural fidelity and a global statistical naturalness measures produce surprisingly accurate predictions of SISR image quality when tested using public subject-rated SISR image datasets. Code of the proposed SFSN model is publicly available at https://***/weizhou-geek/SFSN. Copyright © 2021, The Authors. All rights reserved.

关键词： Generative adversarial networks

A Hybrid and Explainable Deep Learning Framework for SAR Images

学校读者我要写书评

暂无评论

A Hybrid and Explainable Deep Learning Framework for SAR Ima...

IEEE International Symposium on Geoscience and Remote Sensing (IGARSS)

作者： Zhongling Huang Mihai Datcu Zongxu Pan Bin Lei Key Laboratory of Technology in Geo-spatial Information Processing and Application System CAS Remote Sensing Technology Institute (IMF) German Aerospace Center (DLR)

ISBN: (数字)9781728163741

ISBN: (纸本)9781728163758

Deep learning based patch-wise Synthetic Aperture Radar (SAR) image classification usually requires a large number of labeled data for training. Aiming at understanding SAR images with very limited annotation and taking full advantage of complex-valued SAR data, this paper proposes a general and practical framework for quad-, dual-, and single-polarized SAR data. In this framework, two important elements are taken into consideration: image representation and physical scattering properties. Firstly, a convolutional neural network is applied for SAR image representation. Based on time-frequency analysis and polarimetric decomposition, the scattering labels are extracted from complex SAR data with unsupervised deep learning. Then, a bag of scattering topics for a patch is obtained via topic modeling. By assuming that the generated scattering topics can be regarded as the abstract attributes of SAR images, we propose a soft constraint between scattering topics and image representations to refine the network. Finally, a classifier for land cover and land use semantic labels can be learned with only a few annotated samples. The framework is hybrid for the combination of deep neural network and explainable approaches. Experiments are conducted on Gaofen-3 complex SAR data and the results demonstrate the effectiveness of our proposed framework.

关键词： Deep learning Training Annotations Scattering Image representation Radar polarimetry Synthetic aperture radar

Multi-Angular SAR Statistical Properties Analysis and Man-Made Target Detection

学校读者我要写书评

暂无评论

Multi-Angular SAR Statistical Properties Analysis and Man-Ma...

IEEE International Symposium on Geoscience and Remote Sensing (IGARSS)

作者： Fei Teng Yun Lin Yanping Wang Wenjie Shen Shanshan Feng Wen Hong Key Laboratory of Technology in Geospatial Information Processing and Application System Chinese Academy of Sciences Beijing China School of Electronic Information Engineering North China University of Technology Beijing China

ISBN: (数字)9781728163741

ISBN: (纸本)9781728163758

In conventional synthetic aperture radar (SAR) working mode, targets are assumed isotropic due to the limited aperture length. However, most of man-made targets are anisotropic. Therefore, the anisotropic scattering can help us do man-made target detection. Circular SAR (CSAR) [1] is a new SAR working mode and it can obtain the anisotropic scattering of the target by 360° observation. In this paper, the multi-angular statistical properties of targets are analyzed. The probability density functions (PDF) of the anisotropic target are various under different aspect viewing angles, while the PDFs of the isotropic target are basically stable. Then a man-made target detection method is proposed based on the multi-angular statistical property. Likelihood ratio test [2] is used to judge whether the statistical property of scattering is anisotropic or isotropic. Then anisotropic scatterings, which represent the man-made targets, can be discriminated from isotropic scatterings by thresholding. An X-band chamber circular SAR data and a C-band airborne circular SAR data are used to illustrated our idea.

关键词： Scattering Radar polarimetry Synthetic aperture radar Object detection Apertures Fitting Buildings

Sasl: Saliency-adaptive sparsity learning for neural network acceleration

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Shi, Jun Xu, Jianfeng Tasaka, Kazuyuki Chen, Zhibo CAS Key Laboratory of Technology in Geo-spatial Information Processing and Application System University of Science and Technology of China Hefei230027 China KDDI Research Japan

Accelerating the inference speed of CNNs is critical to their deployment in real-world applications. Among all the pruning approaches, those implementing a sparsity learning framework have shown to be effective as they learn and prune the models in an end-to-end data-driven manner. However, these works impose the same sparsity regularization on all filters indiscriminately, which can hardly result in an optimal structuresparse network. In this paper, we propose a Saliency-Adaptive Sparsity Learning (SASL) approach for further optimization. A novel and effective estimation of each filter, i.e., saliency, is designed, which is measured from two aspects: The importance for the prediction performance and the consumed computational resources. During sparsity learning, the regularization strength is adjusted according to the saliency, so our optimized format can better preserve the prediction performance while zeroing out more computation-heavy filters. The calculation for saliency introduces minimum overhead to the training process, which means our SASL is very efficient. During the pruning phase, in order to optimize the proposed data-dependent criterion, a hard sample mining strategy is utilized, which shows higher effectiveness and efficiency. Extensive experiments demonstrate the superior performance of our method. Notably, on ILSVRC-2012 dataset, our approach can reduce 49.7% FLOPs of ResNet-50 with very negligible 0.39% top-1 and 0.05% top-5 accuracy degradation. Copyright © 2020, The Authors. All rights reserved.

关键词： Convolutional neural networks

Vibration Error Compensation Imaging Algorithm for Arc Array SAR Besed on Parameter Estimation 6

学校读者我要写书评

暂无评论

Vibration Error Compensation Imaging Algorithm for Arc Array...

6th Asia-Pacific Conference on Synthetic Aperture Radar, APSAR 2019

作者： Dong, Yifan Wu, Congfei Xu, Wei Huang, Pingping Tan, Weixian Hong, Wen College of Information Engineering Inner Mongolia University of Technology China Inner Mongolia Key Laboratory of Radar Technology and Application Hohhot China Institute of Electronics Chinese Academy of Sciences China Key Laboratory of Technology in Geospatial Information Processing and Application System Beijing China

ISBN: (纸本)9781728129129

Arc array synthetic aperture radar (SAR) is a novel array imaging system for wide-area observation, with wide observation range and high resolution. Arc array synthetic aperture radar uses W-band as carrier signal, and W-band wavelength is short. The small high-frequency vibration of the helicopter-borne platform will cause significant changes in the phase of the echo signal, which will seriously deteriorate the imaging performance of arc array synthetic aperture radar. Based on the high order approximate imaging algorithm of arc array radar, this paper proposes a vibration phase error compensation imaging algorithm based on Short-time Fourier transform (STFT) parameter estimation. The algorithm uses the high-order approximation imaging algorithm of curved array radar to perform high-order approximation of the slant range model, compensates the range cell migration (RCM) and range azimuth coupling in the two-dimensional frequency domain, and finally realizes the vibration error compensation in the range Doppler domain. The simulation results of the lattice target verify the effectiveness of the imaging method. © 2019 IEEE.

关键词： Synthetic aperture radar