分析了Criminisi等人提出的基于样本块的图像修复算法,针对其不足,提出一种改进的基于样本块的快速图像修复方法.引入新的度量函数更新置信度,使优先级的计算更加准确;待匹配块的再筛选策略降低了选择最佳匹配块的随机性;已修复样本块邻域检测避免了全局范围内寻找破损边缘.实验结果表明:该方法取得了较好的修复效果,同时提高了算法的效率.%In view of the deficiency in exemplar-based image inpainting algorithm, proposed by Criminisi and later researcher, a new exemplar-based image inpainting fast algorithm is presented. The new method is introduced to compute confidence term when a damaged block is repaired , which can improve the priori computation precision. Re-selection strategy is adopted to avoid random selection of optimal exemplar. Furthermore, a damaged front searching method within the adjacent region of repaired block is used to narrow the search area of the sample graph to improve the search precision and efficiency. Experimental results show that the proposed method achieves better inpainting effect and efficiency and saves the need of the damaged front selection.
展开▼