首页 > 论文 > 激光与光电子学进展 > 56卷 > 1期(pp:11008--1)

融合特征和决策的卷积-反卷积图像分割模型

Convolution-Deconvolution Image Segmentation Model for Fusion Features and Decision

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

基于全卷积网络提出了一种图像分割模型以获取目标分割结果, 模型包含两个结构相同的深层神经网络分支, 每个分支采用卷积-反卷积的结构实现特征提取和从特征恢复目标区域; 两个分支接收不同类型图像输入, 将来源于两个分支的结果通过加权融合得到最终的分割结果。模型融合了不同图像源的多级尺度特征, 在训练样本数有限的情况下, 通过数据增强使训练得到的模型稳健性更强。在光学图像数据集Weizmann horse和遥感影像数据集Vaihigen上进行实验, 并与相关文献进行比较, 结果表明, 所提模型具有更高的目标分割完整度和最优的分割性能, 在训练数据有限、形态各异、尺度变化较大等的遥感影像建筑物提取中取得了理想的结果, 表明该模型可应用于复杂的遥感影像目标分割。

Abstract

Based on the full convolutional network, an image segmentation model is proposed to obtain the target segmentation results. This model consists of two deep neural network branches with the same structures. As for each branch, a convolution-deconvolution structure is adopted to implement the feature extraction and to recover the target area from the features. These two branches receive different image inputs, and then the final segmentation results are obtained via the weighted fusion of the results from these two branches. This model combines the multi-level scale features of different image sources, and the training model is more robust through data enhancement when the number of training samples is limited. The experiments are carried out on the optical image dataset of Weizmann horse and the remote sensing image dataset of Vaihigen. The comparison with the related literatures is also made. The results show that the proposed model has a higher target segmentation integrity and an optimal segmentation performance. The ideal extraction results of remote sensing image buildings under the conditions of limited training data, various shapes, large scale changes and so on, indicate that the proposed model can be applied to the complex remote sensing image object segmentation.

Newport宣传-MKS新实验室计划
补充资料

中图分类号:O436

DOI:10.3788/lop56.011008

所属栏目:图像处理

基金项目:国家自然科学基金(41471280, 61401265, 61701290, 61701289)

收稿日期:2018-08-08

修改稿日期:2018-09-12

网络出版日期:2018-09-18

作者单位    点击查看

冯晨霄:陕西师范大学计算机科学学院, 陕西 西安 710119
汪西莉:陕西师范大学计算机科学学院, 陕西 西安 710119陕西师范大学现代教学技术教育部重点实验室, 陕西 西安 710119

联系人作者:汪西莉(wangxili@snnu.edu.cn)

【1】Long J, Shelhamer E, Darrell T. Fully convolutional networks for semantic segmentation[C]∥IEEE Conference on Computer Vision and Pattern Recognition(CVPR), 2015: 3431-3440.

【2】Liu W, Rabinovich A, Berg A C. Parsenet: looking wider to see better[J]. arXiv, 2015: 1506.04579.

【3】Wang P Q, Chen P F, Yuan Y, et al. Understanding convolution for semantic segmentation[C]∥IEEE Winter Conference on Applications of Computer Vision (WACV), 2018: 1451-1460.

【4】Noh H, Hong S, Han B. Learning deconvolution network for semantic segmentation[C]∥IEEE International Conference on Computer Vision(ICCV), 2015: 1520-1528.

【5】Ghiasi G, Fowlkes C C. Laplacianpyramid reconstruction and refinement for semantic segmentation[C]∥European Conference on Computer Vision. Springer, 2016:519-534.

【6】Zhao H S, Shi J P, Qi X J, et al. Pyramid scene parsing network[C]∥IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017: 6230-6239.

【7】Dai J F, Qi H Z, Xiong Y W, et al. Deformable convolutional networks[C]∥IEEE International Conference on Computer Vision (ICCV), 2017: 764-773.

【8】Zhang R, Tang S, Zhang Y D, et al. Scale-adaptive convolutions for scene parsing[C]∥IEEE International Conference on Computer Vision (ICCV), 2017: 2050-2058.

【9】Liu F, Liu P Y, Zhang J N, et al. Joint detection of RGB-D images based on double flow convolutional neural network[J]. Laser & Optoelectronics Progress, 2018, 55(2): 021503.
刘帆, 刘鹏远, 张峻宁, 等. 基于双流卷积神经网络的RGB-D图像联合检测[J]. 激光与光电子学进展, 2018, 55(2): 021503.

【10】Li Q W, Zhou Y Q, Ma Y P, et al. Salient object detection method based on binocular vision[J]. Acta Optica Sinica, 2018, 38(3): 0315002.
李庆武, 周亚琴, 马云鹏, 等. 基于双目视觉的显著性目标检测方法[J]. 光学学报, 2018, 38(3): 0315002.

【11】Xu Y, Sun M S.Convolution neural network image defogging based on multi-feature fusion[J]. Laser & Optoelectronics Progress, 2018, 55(3): 031012.
徐岩, 孙美双. 基于多特征融合的卷积神经网络图像去雾算法[J]. 激光与光电子学进展, 2018, 55(3): 031012.

【12】Borenstein E, Ullman S. Combined top-down/bottom-up segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008, 30(12): 2109-2125.

【13】Gerke M. Use of the stair vision library within the ISPRS 2D semantic labeling benchmark (vaihingen)[J]. Technical Report, University of Twente, 2014.

【14】Heipke C, Mayer H, Wiedemann C, et al. Evaluation of automatic road extraction[J]. International Archives of Photogrammetry and Remote Sensing, 1997, 32(3 SECT 4W2): 151-160.

【15】Jia Y, Shelhamer E, Donahue J, et al. Caffe: convolutional architecture for fast feature embedding[C]∥Proceedings of the 22nd ACM International Conference on Multimedia. ACM, 2014: 675-678.

【16】Bertelli L, Yu T, Vu D, et al. Kernelized structural SVM learning for supervised object segmentation[C]∥Computer Vision and Pattern Recognition, 2011: 2153-2160.

【17】Yang J, Price B, Cohen S, et al. Patchcut: data-driven object segmentation via local shape transfer[C]∥Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2015: 1770-1778.

【18】Liu F Y, Lin G S, Shen C H. CRF learning with CNN features for image segmentation[J]. Pattern Recognition, 2015, 48(10): 2983-2992.

【19】Visin F, Ciccone M, Romero A, et al. Reseg: a recurrent neural network-based model for semantic segmentation[C]∥Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. 2016: 41-48.

【20】Badrinarayanan V, Kendall A, Cipolla R. SegNet: a deep convolutional encoder-decoder architecture for image segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(12): 2481-2495.

【21】Paisitkriangkrai S, Sherrah J, Janney P, et al. Effective semantic pixel labelling with convolutional networks and conditional random fields[C]∥Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. 2015: 36-43.

【22】Audebert N, Le Saux B, Lefèvre S. Semantic segmentation of earth observation data using multimodal and multi-scale deep networks[C]∥Asian Conference on Computer Vision. Cham: Springer, 2016: 180-196.

引用该论文

Feng Chenxiao,Wang Xili. Convolution-Deconvolution Image Segmentation Model for Fusion Features and Decision[J]. Laser & Optoelectronics Progress, 2019, 56(1): 011008

冯晨霄,汪西莉. 融合特征和决策的卷积-反卷积图像分割模型[J]. 激光与光电子学进展, 2019, 56(1): 011008

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF