摘要
通过对块状缺损图像修复技术的研究,提出一种改进的纹理合成修复算法。继而依据粗糙度的分析对图像块进行分类,根据修复算法特性人为丢弃部分图像块,并结合JPEG算法进行压缩编码。在相近的视觉效果下,可以使图像压缩率得到显著的提高。
An optimized approach based on texture synthesis for filling-in blocks of missing data is presented in this paper According to coarseness analysis and the character of inpainting algorithm, parts of image blocks would be dropped purposively. And then, this method is applied to the second-time image compression along with other compression algorithms such as JPEG, and the compression ratio is highly improved.
出处
《电子与信息学报》
EI
CSCD
北大核心
2006年第5期848-851,共4页
Journal of Electronics & Information Technology
关键词
图像修复
纹理合成
粗糙度
JPEG
压缩率
Image inpainting, Texture synthesis, Coarseness, JPEG, Compression ratio
