帮助 关于我们

返回检索结果

基于全卷积神经网络的非对称并行语义分割模型
Asymmetric Parallel Semantic Segmentation Model Based on Full Convolutional Neural Network

查看参考文献22篇

文摘 针对RGB图像具有丰富的色彩细节特征,红外图像对目标轮廓、尺寸、边界等外形特征有较高敏感度的特点,提出了一种非对称并行语义分割模型APFCN( Asymmetric Parallelism Fully Convolutional Networks). APFCN上路设计了一个卷积核尺寸非统一的五层空洞卷积网络来提取红外图像目标高层轮廓特征;下路沿用卷积加池化网络提取RGB图像三个尺度上的细节特征;后端将红外图像高层特征与RGB图像三个尺度的细节特征进行融合,并将4倍上采样后的融合特征作为语义分割输出.结果表明,APFCN在像素精度和交并比等方面均优于FCN(输入为RGB图像或红外图像) ,适用于背景一致下地面目标的语义分割任务.
其他语种文摘 Aiming at that RGB image is rich in color details of scene and infrared image is sensitive to outline、size and boundary of target,a novel semantic segmentation model APFCN ( Asymmetric Parallelism Fully Convolutional Networks) is proposed. In the upper part of APFCN,a five layer dilation convolution network,where the five kernel sizes are not uniform, is designed used to extract the high-level targets contour features of infrared image. In the lower part of APFCN, aclassical CNN network is used to extract three scale features of RGB images. At the back of APFCN, the high level features of the infrared image are fused with the three scale features of the RGB image, and the fused features after 4 times upper sampling is used as the semantic segmentation output of APFCN. The results show that APFCN is better than FCN ( input RGB image or infrared image) in PA ( Pixel Accuracy) and MIoU ( Mean Intersection over Union). APFCN is suitable for the semantic segmentation task of ground targets with consistent background colors.
来源 电子学报 ,2019,47(5):1058-1064 【核心库】
DOI 10.3969/j.issn.0372-2112.2019.05.012
关键词 语义分割 ; 全卷积神经网络 ; 非对称并行全卷积神经网络 ; 空洞卷积 ; 空洞率
地址

西北工业大学航海学院, 陕西, 西安, 710072

语种 中文
文献类型 研究性论文
ISSN 0372-2112
学科 自动化技术、计算机技术
基金 国家自然科学基金
文献收藏号 CSCD:6668526

参考文献 共 22 共2页

1.  He X. Multiscale conditional random fields for image labeling. Proceedings of Computer Vision and Pattern Recognition,2004:695-702 CSCD被引 1 次    
2.  Yang L. Multiple class segmentation using A unified framework over mean-shift patches. Proceedings of Computer Vision and Pattern Recognition,2007:1-8 CSCD被引 1 次    
3.  Pantofaru C. Object recognition by integrating multiple image segmentations. Proceedings of European Conference on Computer Vision,2008:481-494 CSCD被引 2 次    
4.  Gould S. Decomposing a scene into geometric and semantically consistent regions. Proceedings of International Conference on Computer Vision,2009:1-8 CSCD被引 2 次    
5.  Kumar M P. Efficiently selecting regions for scene understanding. Proceedings of Computer Vision and Pattern Recognition,2010:3217-3224 CSCD被引 1 次    
6.  Jain A. Piecing together the segmentation jigsaw using context. Proceedings of Computer Vision and Pattern Recognition,2011:2001-2008 CSCD被引 1 次    
7.  Kohli P. Robust higher order potentials for enforcing label consistency. Proceedings of Computer Vision and Pattern Recognition,2008:1-8 CSCD被引 1 次    
8.  Russell C. Associative hierarchical CRFs for object class image segmentation. Proceedings of International Conference on Computer Vision,2009:739-746 CSCD被引 1 次    
9.  Ladicky L. Graph cut based inference with co-occurrence statistics. Proceedings of European Conference on Computer Vision,2010:239-253 CSCD被引 1 次    
10.  Hinton G. Where do features come from ?. Cognitive Science,2014,38(6):1078-1101 CSCD被引 7 次    
11.  Lecun Y. Deep learning. Nature,2015,521(7553):436-444 CSCD被引 3268 次    
12.  Krizhevsky A. ImageNet classification with deep convolutional neural networks. Proceedings of Neural Information Processing Systems,2012:1097-1105 CSCD被引 6 次    
13.  Szegedy C. Going deeper with convolutions. Proceedings of Computer Vision and Pattern Recognition,2015:1-9 CSCD被引 3 次    
14.  Farabet C. Learning hierarchical features for scene labeling. IEEE Transactions on Pattern Analysis and Machine Intelligence,2013,35(8):1915-1929 CSCD被引 142 次    
15.  Long J. Fully convolutional networks for semantic segmentation. Proceedings of Computer Vision and Pattern Recognition,2015:3431-3440 CSCD被引 2 次    
16.  Long J. Fully convolutional networks for semantic segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence,2014,39(4):640-651 CSCD被引 139 次    
17.  Yu F. Multi-scale context aggregation by dilated convolutions. Proceedings of International Conference on Learning Representations,2016 CSCD被引 4 次    
18.  Chen L C. DeepLab: Semantic image segmentation with deep convolutional nets, atrous convolution,and fully connected CRFs. IEEE Transactions on Pattern Analysis and Machine Intelligence,2018,40(4):834-848 CSCD被引 776 次    
19.  Wang P. Understanding convolution for semantic segmentation. Proceedings of Computer Vision and Pattern Recognition,2018:1451-1460 CSCD被引 1 次    
20.  周则明. 基于流形特征与形状先验的红外直升机图像分割. 电子学报,2018,46(4):834-839 CSCD被引 1 次    
引证文献 8 篇

1 罗会兰 基于深度网络的图像语义分割综述 电子学报,2019,47(10):2211-2220
CSCD被引 17 次

2 徐少平 一种非开关型快速随机脉冲噪声降噪算法 电子学报,2019,47(12):2622-2629
CSCD被引 1 次

显示所有8篇文献

论文科学数据集
PlumX Metrics
相关文献

 作者相关
 关键词相关
 参考文献相关

版权所有 ©2008 中国科学院文献情报中心 制作维护：中国科学院文献情报中心
地址：北京中关村北四环西路33号 邮政编码：100190 联系电话：(010)82627496 E-mail:cscd@mail.las.ac.cn 京ICP备05002861号-4 | 京公网安备11010802043238号