作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2006, Vol. 32 ›› Issue (11): 206-208.

• 人工智能及识别技术 • 上一篇    下一篇

基于块缺失图像修复技术的研究与应用

王 晨,杜 彦,杜建洪   

  1. 复旦大学信息学院,上海 200433
  • 出版日期:2006-06-05 发布日期:2006-06-05

Research and Application of Image Inpainting Based on Missing Image Blocks

WANG Chen,DU Yan DU Jianhong   

  1. School of Information, Fudan University, Shanghai 200433
  • Online:2006-06-05 Published:2006-06-05

摘要: 图像修复是数字图像处理的重要内容,在图像润色、视频图像复原等诸多领域都有着非常广泛的应用。该文以对图像块粗糙度的分析为依据,优化纹理合成算法,并结合JPEG 算法,应用于图像压缩,使图像压缩率得到显著提高。

关键词: 图像修复;纹理合成;压缩;粗糙度

Abstract: This paper presents an optimized approach to texture synthesis based on coarseness analysis, which reconstructs the lost image blocks with the information abstracted from the neighborhood. 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.

Key words: Image inpainting; Texture synthesis; Compression; Coarseness