Skeleton extraction via structure-adaptive anisotropic filtering | |
Du, Xiaofeng ; Zhu, Shunzhi ; Li, Jianmin ; Du XF(杜晓凤) | |
2014 | |
关键词 | Anisotropy Extraction Internet Object recognition |
英文摘要 | Conference Name:6th International Conference on Internet Multimedia Computing and Service, ICIMCS 2014. Conference Address: Xiamen, China. Time:July 10, 2014 - July 12, 2014.; National Natural Foundation of China; SIGMM China Chapter; Xiamen University; Skeletonization of gray-scale images is a challenging problem in computer vision due to the non-uniform width of shape and the clutter background. This paper presents a novel approach of skeletonization for gray-scale images directly from original image based on anisotropic Gaussian filter. To deal with the non-uniform width of natural object parts, we adapt the shape of filter kernel to local gradient feature. The orientation of filter is firstly estimated based on local structure tensor, and then the scale is calculated based on gradient vector flux. After that, the anisotropic Gaussian filter is performed on the image. The skeleton strength map is defined by the gradient vector flux measure. Finally, thin and binary skeleton is obtained by non- maximum suppression the skeleton strength map. Our method performs well on both binary and gray image in skeleton extraction even for clutter image. Copyright 2014 ACM. |
语种 | 英语 |
出处 | http://dx.doi.org/10.1145/2632856.2632897 |
出版者 | Association for Computing Machinery |
内容类型 | 其他 |
源URL | [http://dspace.xmu.edu.cn/handle/2288/85905] |
专题 | 数学科学-会议论文 |
推荐引用方式 GB/T 7714 | Du, Xiaofeng,Zhu, Shunzhi,Li, Jianmin,et al. Skeleton extraction via structure-adaptive anisotropic filtering. 2014-01-01. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论