Incorporating Prior Shape into Geometric Active Contours for Face Contour Detection

• 作者: HUANGFuzhen SUJianbo XIYugeng
• 来源期刊: 电子学报(英文版)
• 学科: 工学
• 关键词: 表面轮廓检测 几何主动相关 水平集 图像处理
• 年，卷（期）: 2004,（3）
• 页码范围: 552-555
• 页数: 4页
• 分类号: TP391.41
• 语种: 中文

摘要

In this paper a new method that incorporates prior shape information into geometric active contours for face contour detection is proposed. As in general a human face can be treated as an ellipse with a little shape variation, the prior face shape is represented as an elliptical curve. By combining the prior face shape with the powerful geometric active model proposed by Chan and Vese, the improved geometric active model can retain all the advantage of the Chan-Vese model and can detect face contours in images with complex backgrounds accurately even if the image is noisy. Moreover, by implementing the new model in a variational level set framework, automatic topological changes of the model can be achieved naturally and the transformation parameters that map the face boundary to the prior shape can be roughly estimated simultaneously. The experimental results show our procedure to be eiTicient.