Semantic segmentation based on DeepLabV3+ and superpixel optimization

doi:10.3788/OPE.20192712.2722

CIOMP OpenIR

	Semantic segmentation based on DeepLabV3+ and superpixel optimization
	F.-L.Ren; X.He; Z.-H.Wei; Y.Lu; M.-Y.Li
	2019
发表期刊	Guangxue Jingmi Gongcheng/Optics and Precision Engineering
ISSN	1004924X
卷号	27 期号:12 页码:2722-2729
摘要	To tackle the problem where by DeepLabV3+ loses considerable detail information during feature extraction, which leads to poor segmentation results in the edges of the objects, this study proposed a semantics segmentation algorithm based on DeepLabV3+ and optimized by superpixels. First, a DeepLabV3+ model was chosen to extract semantic features and obtain coarse semantic segmentation results. Then, the simple linear iterative clustering algorithm was used to segment the input image into superpixels. Finally, high-level abstract semantic features and detailed information of the superpixels were fused to obtain edge optimized semantic segmentation results. Experiments conducted on the PASCAL VOC 2O12 dataset show that compared to DeepLabV3+, the proposed algorithm had superior performance in terms of detail parts such as edges of objects, and the value of mIoU reached 83.8%.The proposed algorithm thus outperformed other state-of-the-art algorithms in terms of semantic segmentation. 2019, Science Press. All right reserved.
关键词	Image segmentation,Clustering algorithms,Deep learning,Iterative methods,Semantics,Superpixels
DOI	10.3788/OPE.20192712.2722
URL	查看原文
收录类别	EI
引用统计
文献类型	期刊论文
条目标识符	http://ir.ciomp.ac.cn/handle/181722/63094
专题	中国科学院长春光学精密机械与物理研究所
推荐引用方式 GB/T 7714	F.-L.Ren,X.He,Z.-H.Wei,et al. Semantic segmentation based on DeepLabV3+ and superpixel optimization[J]. Guangxue Jingmi Gongcheng/Optics and Precision Engineering,2019,27(12):2722-2729.
APA	F.-L.Ren,X.He,Z.-H.Wei,Y.Lu,&M.-Y.Li.(2019).Semantic segmentation based on DeepLabV3+ and superpixel optimization.Guangxue Jingmi Gongcheng/Optics and Precision Engineering,27(12),2722-2729.
MLA	F.-L.Ren,et al."Semantic segmentation based on DeepLabV3+ and superpixel optimization".Guangxue Jingmi Gongcheng/Optics and Precision Engineering 27.12(2019):2722-2729.

条目包含的文件		下载所有文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
Semantic segmentatio（2102KB）	期刊论文	出版稿	开放获取	CC BY-NC-SA	浏览下载