Ingenious snake: An Adaptive Multi-class Contours Extraction
文献类型：会议论文
| 作者 | Baolin Li; Shoujun Zhou |
| 出版日期 | 2018 |
| 会议日期 | 2018 |
| 会议地点 | 香港 |
| 英文摘要 | Active contour model (ACM) plays an important role in computer vision and medical image application. The traditional ACMs were used to extract single-class of object contours. While, simultaneous extraction of multi-class of interesting contours (i.e., various contours with closed- or open-ended) have not been solved so far. Therefore, a novel ACM model named "Ingenious Snake" is proposed to adaptively extract these interesting contours. In the first place, the ridge-points are extracted based on the local phase measurement of gradient vector flow field; the consequential ridgelines initialization are automated with high speed. Secondly, the contours’ deformation and evolvement are implemented with the ingenious snake. In the experiments, the result from initialization, deformation and evolvement are compared with the existing methods. The quantitative evaluation of the structure extraction is satisfying with respect of effectiveness and accuracy. |
| URL标识 | 查看原文 |
| 源URL | [http://ir.siat.ac.cn:8080/handle/172644/14490]  |
| 专题 | 深圳先进技术研究院_医工所 |
| 推荐引用方式 GB/T 7714 | Baolin Li,Shoujun Zhou. Ingenious snake: An Adaptive Multi-class Contours Extraction[C]. 见:. 香港. 2018. |
入库方式： OAI收割
来源：深圳先进技术研究院
浏览0
下载0
收藏0
其他版本
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。