中文摘要：

图像修复是一种去除图像中多余可视物体并用合理的纹理填充未知区域的技术。针对Criminisi等人提出的基于样本的图像修复算法在寻找匹配块时存在的一些不足,提出一种基于模式相似性的修复算法。该算法使用样本纹理的亮度变化和空域特性作为寻找匹配模块的约束条件。实验结果表明该方法在修复强结构纹理图像时有很好的效果。

英文摘要：

Image inpainting is a technology for removing undesired visual objects in images and filling the missing regions with plausible textures.Since the image inpainting algorithm based on examplar,proposed by Criminisi and his partners,has some deficiencies in determining matching block size,an improved image inpainting algorithm based on pattern similarity is presented in this paper.This algorithm obtains a more accurate match by allowing brightness changes of sample textures and introducing spatial locality as constraint conditions.The experiment results show this method is effective as structured texture images are inpainted.