Improved Active Contour Model with Texture-Contrast Forces
碩士 === 國立嘉義大學 === 資訊工程研究所 === 93 === Active contour model provides a powerful tool to deal with computer vision tasks such as image segmentation, image representation, shape modeling, video-object shape extraction, and medical image. But it heavily depends on image’s local information no matter how...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
2005
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/93652751455012114489 |
| Summary: | 碩士 === 國立嘉義大學 === 資訊工程研究所 === 93 === Active contour model provides a powerful tool to deal with computer vision tasks such as image segmentation, image representation, shape modeling, video-object shape extraction, and medical image. But it heavily depends on image’s local information no matter how inappropriate the information is. Besides, in general, snake models are very sensitive to position initialization. In order to improve the Active contour model to fit the image object better. We adopt the texture analysis to understand the image information. We also use the textural contrast force to solve the drawback of traditional Active contour model and leave the object’s concavity to be well fitted.
|
|---|