计算机工程与应用 ›› 2015, Vol. 51 ›› Issue (19): 206-209.

• 图形图像处理 • 上一篇    下一篇

轮廓波系数引导的数字图像修复算法研究

李宣妮,王慧琴,吴  萌   

  1. 西安建筑科技大学 信息与控制工程学院,西安 710055
  • 出版日期:2015-09-30 发布日期:2015-10-13

Research of image inpainting algorithm led by contourlet coefficients

LI Xuanni, WANG Huiqin, WU Meng   

  1. School of Information and Control Engineering, Xi’an University of Architecture and Technology, Xi’an 710055, China
  • Online:2015-09-30 Published:2015-10-13

摘要: 为提高数字图像修复的精度,通过提取图像的轮廓波系数,用轮廓波系数分析图像的纹理信息,改进了基于样本的图像修复算法。引入了[β]因子评价图像的纹理信息,改进了优先值计算公式,优化了图像的修复顺序,使纹理复杂度较高的部分被先修复。实验结果表明,该方法有效避免了图像修复产生的图像模糊和马赛克效应,同时降低了算法的时间复杂度。

关键词: 轮廓波, 图像修复, 基于样本, 优先级

Abstract: In order to improve the accuracy of image inpainting, this paper extracts the contourlet coefficients of the original image, and then uses the contourlet coefficients to analyze texture information of the image, improves the priority calculation formula. By introducing a factor β to estimate texture information, it optimizes the inpainting order, repairs the complex vein parts first, and improves the sample-based image inpainting algorithm. The experimental results show that this method can avoid image blur and mosaic effect effectively, and reduce the time complexity of the algorithm.

Key words: contourlet, image inpainting, sample-based, priority