中文摘要：

纹理合成是图像修补采用的主流方法，提出一种新的结构自适应纹理合成算法以提高纹理合成质量．首先通过分析图像灰度共生矩阵和灰度行程矩阵的特征参数确定纹理类型，然后对于显著结构纹理图像采用基于结构相似度的匹配准则以提高合成质量．实验结果表明本文提出的自适应算法可以有效提高显著结构纹理图像的合成质量．

英文摘要：

Texture Synthesis is the major method of image completion. We introduce a novel structure adaptive algorithm to improve the quality of texture synthesis. Firstly, we analyze texture features from gray co-occurrence matrix and gray run-length matrix to fmd the texture type. Then we choose the patch-matching method based on structure similarity for salient structure texture images, thus im- prove texture synthesis quality. Experimental results demonstrate that the proposed algorithm can improve texture synthesis quality effectively.