检索结果-内蒙古大学图书馆

W-net: Simultaneous segmentation of multi-anatomical retinal structures using a multi-task deep neural network

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Zhao, Hongwei Peng, Chengtao Liu, Lei Li, Bin School of Information Science and Technology University of Science and Technology of China Hefei Anhui230022 China Department of Precision Machinery and Instrumentation University of Science and Technology of China HefeiAnhui230022 China CAS Key Laboratory of Technology in Geo-spatial Information Processing and Application System University of Science and Technology of China Hefei Anhui230026 China

Segmentation of multiple anatomical structures is of great importance in medical image analysis. In this study, we proposed a W-net to simultaneously segment both the optic disc (OD) and the exudates in retinal images based on the multi-task learning (MTL) scheme. We introduced a class-balanced loss and a multi-task weighted loss to alleviate the imbalanced problem and to improve the robustness and generalization property of the W-net. We demonstrated the effectiveness of our approach by applying five-fold cross-validation experiments on two public datasets e_ophtha_EX and DiaRetDb1. We achieved F1-score of 94.76% and 95.73% for OD segmentation, and 92.80% and 94.14% for exudates segmentation. To further prove the generalization property of the proposed method, we applied the trained model on the DRIONS-DB dataset for OD segmentation and on the MESSIDOR dataset for exudate segmentation. Our results demonstrated that by choosing the optimal weights of each task, the MTL based W-net outperformed separate models trained individually on each task. Code and pre-trained models will be available at: https://***/FundusResearch/MTL_for_OD_and_***. Copyright © 2020, The Authors. All rights reserved.

关键词： Deep neural networks

Stereoscopic omnidirectional image quality assessment based on predictive coding theory

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Chen, Zhibo Xu, Jiahua Lin, Chaoyi Zhou, Wei CAS Key Laboratory of Technology in Geo-Spatial Information Processing and Application System University of Science and Technology of China Hefei Anhui230027 China

Objective quality assessment of stereoscopic omnidirectional images is a challenging problem since it is influenced by multiple aspects such as projection deformation, field of view (FoV) range, binocular vision, visual comfort, etc. Existing studies show that classic 2D or 3D image quality assessment (IQA) metrics are not able to perform well for stereoscopic omnidirectional images. However, very few research works have focused on evaluating the perceptual visual quality of omnidirectional images, especially for stereoscopic omnidirectional images. In this paper, based on the predictive coding theory of the human vision system (HVS), we propose a stereoscopic omnidirectional image quality evaluator (SOIQE) to cope with the characteristics of 3D 360-degree images. Two modules are involved in SOIQE: predictive coding theory based binocular rivalry module and multi-view fusion module. In the binocular rivalry module, we introduce predictive coding theory to simulate the competition between high-level patterns and calculate the similarity and rivalry dominance to obtain the quality scores of viewport images. Moreover, we develop the multi-view fusion module to aggregate the quality scores of viewport images with the help of both content weight and location weight. The proposed SOIQE is a parametric model without necessary of regression learning, which ensures its interpretability and generalization performance. Experimental results on our published stereoscopic omnidirectional image quality assessment database (SOLID) demonstrate that our proposed SOIQE method outperforms state-of-the-art metrics. Furthermore, we also verify the effectiveness of each proposed module on both public stereoscopic image datasets and panoramic image datasets. Copyright © 2019, The Authors. All rights reserved.

关键词： Binocular vision

Online filter clustering and pruning for efficient convnets

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Zhou, Zhengguang Zhou, Wengang Hong, Richang Li, Houqiang CAS Key Laboratory of Technology in Geo-spatial Information Processing and Application System EEIS Department University of Science and Technology of China HeFei University of Technology

Pruning filters is an effective method for accelerating deep neural networks (DNNs), but most existing approaches prune filters on a pre-trained network directly which limits in acceleration. Although each filter has its own effect in DNNs, but if two filters are same with each other, we could prune one safely. In this paper, we add an extra cluster loss term in the loss function which can force filters in each cluster to be similar online. After training, we keep one filter in each cluster and prune others and fine-tune the pruned network to compensate the loss. Particularly, the clusters in every layer can be defined firstly which is effective for pruning DNNs within residual blocks. Extensive experiments on CIFAR10 and CIFAR100 benchmarks demonstrate the competitive performance of our proposed filter pruning method. Copyright © 2019, The Authors. All rights reserved.

关键词： Deep neural networks

DRBOX FAMILY: A GROUP OF OBJECT DETECTION TECHNIQUES FOR REMOTE SENSING IMAGES

学校读者我要写书评

暂无评论

DRBOX FAMILY: A GROUP OF OBJECT DETECTION TECHNIQUES FOR REM...

IEEE International geoscience and Remote Sensing Symposium

作者： Lei Liu Zongxu Pan Guowei Chen Yizhao Gao Key Laboratory of Technology in Geo-spatial Information Processing and Application System Chinese Academy of Sciences Beijing China

Objects in remote sensing images are difficult to detect due to arbitrarily rotated angles and the wide variance of scales. As bounding box plays an important role in object detection, we proposes a new bounding box type named rotated bounding box (rBox). With the application of rBox, we have proposed a series of detection techniques (DrBox, DrBoxLight, DrBoxSemi, DrBoxPro) to effectively handle the situation where the orientation angles of the objects are arbitrary. This article is a brief overview of these techniques. The original DrBox detector applies VGG-net as its main network framework, with image pyramid input to address multi-scale problem. DrBoxLight is a mini version of DrBox, which applies MobileNet and knowledge distillation to be deployed on embedded devices. DrBoxSemi is a semi-supervised version of DrBox, so annotation of all training samples is no longer necessary. DrBoxPro is the most important update for DrBox with professional designing of abundant prior-rBoxes on feature pyramid networks. In our experiments, we demonstrated how rBox helps to improve the performance of object detection compared with traditional bounding boxes. Besides, we evaluated the performance of our DrBox family on a series of object detection tasks.

关键词： Object detection Remote sensing images Rotated object Deep learning

Automatic Registration of Optical and SAR Images VIA Improved Phase Congruency

学校读者我要写书评

暂无评论

Automatic Registration of Optical and SAR Images VIA Improve...

IEEE International Symposium on geoscience and Remote Sensing (IGARSS)

作者： Yuming Xiang Rongshu Tao Feng Wang Hongjian You Key Laboratory of Technology in Geo-spatial Information Processing and Application System Chinese Academy of Sciences Beijing China

Owing to the property of being constant to image contrast and the identification of various types of features, phase congruency (PC) model has been widely used in remote sensing applications. However, when the PC is directly applied to optical and synthetic aperture radar (SAR) image registration, it fails to handle large radiometric and geometric differences. In this paper, we propose an automatic algorithm to solve this problem. First, evenly-distributed keypoints are extracted from the optical images via the block harris method. Complementary grid points are selected in image regions with poor structure and texture information. Then a robust similarity metric based on the improved PC model is proposed. Since the two images show diverse properties, we utilize two different PC models, the traditional PC and the SAR-PC. The PC values of several directions are aggregated to construct the feature descriptors on the basis of which, as a result, a similarity metric using the normalized correlation coefficient (NCC) is obtained. We compare the proposed metric with two baselines (mutual information and NCC) and a state-of-the-art method (histogram of the oriented phase congruency, HOPC) in the case of various scenarios, the results show that our method outperforms the baselines and show comparable performance with HOPC in regions with abundant structure information and better performance in untextured regions.

关键词： Optical imaging Adaptive optics Radar polarimetry Optical sensors Measurement Feature extraction Image edge detection

Object-based method for optical and SAR images change detection

学校读者我要写书评

暂无评论

Object-based method for optical and SAR images change detect...

IET International Radar Conference 2018, IRC 2018

作者： Wan, Ling Zhang, Tao You, Hongjian Key Laboratory of Technology in Geo-spatial Information Processing and Application System Beijing100190 China Institute of Electronics Chinese Academy of Sciences Beijing100190 China University of Chinese Academy of Sciences Beijing100039 China

This study introduces an automatic method for change detection of multi-sensor remote-sensing images (e.g. optical and synthetic aperture radar (SAR) images). As object-based image analysis can effectively reduce the spurious changes and the sensitivity to registration, first, multi-date segmentation is employed to generate homogeneous image objects in spectral, spatial, and temporal, in order to weak the intensity variation effects of multi-sensor images. Then, modified fuzzy c-means (FCM) algorithms are employed to preliminarily classify optical and SAR images, and a criterion is defined using membership values of parcels to select the sample parcels for each class and image. Finally, a change detection principle, which takes statistical properties as the feature space, is introduced to detect changes between multi-sensor images. The experiment results verify that the proposed method is able to cope with optical and SAR images change detection. © 2019 Institution of Engineering and technology. All rights reserved.

关键词： Change detection

Image-to-image translation with multi-path consistency regularization

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Lin, Jianxin Xia, Yingce Wang, Yijun Qin, Tao Chen, Zhibo CAS Key Laboratory of Technology in Geo-spatial Information Processing and Application System University of Science and Technology of China University of Science and Technology of China Microsoft Research Asia

Image translation across different domains has attracted much attention in both machine learning and computer vision communities. Taking the translation from source domain Ds to target domain Dt as an example, existing algorithms mainly rely on two kinds of loss for training: One is the discrimination loss, which is used to differentiate images generated by the models and natural images;the other is the reconstruction loss, which measures the difference between an original image and the reconstructed version through Ds → Dt → Ds translation. In this work, we introduce a new kind of loss, multi-path consistency loss, which evaluates the differences between direct translation Ds → Dt and indirect translation Ds → Da → Dt with Da as an auxiliary domain, to regularize training. For multi-domain translation (at least, three) which focuses on building translation models between any two domains, at each training iteration, we randomly select three domains, set them respectively as the source, auxiliary and target domains, build the multi-path consistency loss and optimize the network. For two-domain translation, we need to introduce an additional auxiliary domain and construct the multi-path consistency loss. We conduct various experiments to demonstrate the effectiveness of our proposed methods, including face-toface translation, paint-to-photo translation, and deraining/de-noising translation. Copyright © 2019, The Authors. All rights reserved.

关键词： Iterative methods

3D reconstruction and error analysis of multi-view space-borne SAR images under different configurations

学校读者我要写书评

暂无评论

3D reconstruction and error analysis of multi-view space-bor...

IET International Radar Conference 2018, IRC 2018

作者： Wang, Chao Qiu, Xiaolan Li, Fangfang Lei, Bin University of the Chinese Academy of Sciences Beijing China Institute of Electrics Chinese Academy of Sciences Beijing China Key Laboratory of Technology in Geo-spatial Information Processing and Application System Institute of Electrics Chinese Academy of Sciences Beijing China

Synthetic aperture radar (SAR) has a good ability to detect the microwave scattering characteristics of the target and has a good capability of slant range Doppler positioning. Using multi-view SAR images in combination with image matching and positioning equations, the 3D position of the target can be obtained. In the past, multi-view 3D reconstruction mainly used side-looking images with parallel trajectories but different incident angles. That method is not universal for different configurations and lacks analysis of the relationship between solution and parameter error. This study aims at the problem of multi-view SAR 3D reconstruction. The authors establish general 3D reconstruction equations that can be used for non-parallel track and non-side-looking and the solution method is deduced. Based on this, an analysis method of error sensitivity is proposed and the dependence relationship between the reconstruction error and configuration, velocity, position, and slant range is analysed. The correctness of this method is verified by simulation experiments, which provides guidance for selection of configuration and sensor accuracy when applying 3D reconstruction. © 2019 Institution of Engineering and technology. All rights reserved.

关键词： Synthetic aperture radar

A STUDY ON THE FREQUENCY AND AZIMUTH COHERENCE OF HIGH-RESOLUTION SAR IMAGE

学校读者我要写书评

暂无评论

A STUDY ON THE FREQUENCY AND AZIMUTH COHERENCE OF HIGH-RESOL...

IEEE International geoscience and Remote Sensing Symposium

作者： Wenji Xing Xiaolan Qiu Chibiao Ding The Key Laboratory of Technology in Geo-spatial Information Processing and Application System Institute of Electronics Chinese Academy of Sciences Beijing China

High-resolution SAR has large transmitting bandwidth and wide synthetic aperture. How to understand and take advantage of the variation characteristics of SAR scattering characteristics with angle and frequency is a topic that worth studying. This article establishes a coherence matrix of sub-band and sub-aperture SAR images, and analyzes its ability to classify scattering mechanism. Experiments are conducted using the TerraSAR-X high-resolution data of different scenarios, and some meaningful results are got, which may provide some support to the analysis and application of high-resolution SAR data.

关键词： Coherence Entropy Azimuth Radar polarimetry Synthetic aperture radar Image resolution