为改善传统CV（Chan和Vese）分割算法对自然图像分割的效果，将自然图像进行自适应保边分解，获取的卡通分量区域内部具有一致性，区域间存在显著差异且目标轮廓清晰。对卡通分量运用水平集实现目标区域分割。实验结果表明，相对于传统CV分割算法，该算法对自然图像分割的效果较好，分割测评分数较高。%In order to improve the performance of the traditional CV segmentation algorithm for natural image segmentation . Firstly ,a natural image is decomposed with adaptive protected edges .Cartoon component had good regional consistency and had significant differences between the regions with clear edges .Then ,the cartoon component is segmented by level set method .The experimental results show that our algorithm has better segmentation results and higher segmentation evaluation score than tradi-tional CV segmentation algorithm .
展开▼
