计算机应用研究2011，Vol.28Issue(6)：2247-2249,3.DOI:10.3969/j.issn.1001-3695.2011.06.068
基于SIFT特征点匹配的水印图像几何校正算法
New watermarked image geometric correction algorithm based on SIFT feature points matching
摘要
Abstract
In order to achieve the synchronization between watermark detection and watermark embedding, this paper proposed a method that estimated watermarked image geometric distortion factors and corrected it by using SIFT feature points matching.Test image rotation and scaling parameters could be estimated by the difference between the stable and the farther SIFT feature point' s coordinates of it and that of the original image. After revising, the test image translation could be corrected by the change of one the stable SIFT feature point' s coordinates of it and that of the original image. Theoretical analysis and experimental results show that the algorithm has high correctional precision, stable performance, and it can recover watermark synchronization of watermarked image by rotation, scaling and translation attacks effectively, so that the watermark can be correctly detected and extracted.关键词
水印图像/几何攻击/SIFT/参数估计Key words
watermarked image/ geometric attacks/ SIFT( scale invariant feature transform)/ estimate affine factors分类
信息技术与安全科学引用本文复制引用
廖琪男..基于SIFT特征点匹配的水印图像几何校正算法[J].计算机应用研究,2011,28(6):2247-2249,3.基金项目
广西教育厅科研项目(201010LX399) （201010LX399）
