基于LPP和改进SIFT的copy-move篡改检测 |
| |
引用本文: | 李子健,阮秋琦. 基于LPP和改进SIFT的copy-move篡改检测[J]. 信号处理, 2017, 33(4): 589-594. DOI: 10.16798/j.issn.1003-0530.2017.04.020 |
| |
作者姓名: | 李子健 阮秋琦 |
| |
作者单位: | 北京交通大学信息科学研究所 |
| |
基金项目: | 973项目(2012CB316304);国家自然科学基金项目(61471032) |
| |
摘 要: | 图像的复制-粘贴篡改检测是图像篡改检测领域中的重要组成部分。本文基于SIFT算法以及LPP的降维思想,提出了一种新的篡改检测算法。本文在SIFT算法的基础上,使用LPP算法对SIFT算法生成的特征点以及特征向量进行降维。使得传统SIFT算法在实际应用中特征点数目过多、特征向量维数过高等缺陷得到了解决。并使用凝聚型层次聚类算法对相似的特征点进行聚类,完成了对图像复制-粘贴篡改区域的检测。在文章的最后,本文对哥伦比亚大学复制-粘贴图像库里的100张图片进行实验。实验结果表明,不管篡改区域后处理方式是拉伸还是旋转,本文算法都能比传统的SIFT、SURF、PCA-SIFT等算法生成更少的特征点数目和更低的特征向量维度,使得检测效率以及检测正确率得到有效提升。
|
关 键 词: | 篡改检测 复制 粘贴篡改方式 尺度不变特征变换 局部保持投影 |
收稿时间: | 2016-11-29 |
Copy-move Forgery Detection Based on LPP and Improved SIFT Algorithm |
| |
Affiliation: | School of Computer and Information Technology,Beijing Jiaotong University |
| |
Abstract: | Image copy-move forgery detection is an important part of image forgery detection. In this paper, we propose a novel forgery detection algorithm based on the SIFT and LPP algorithm. First,we use LPP algorithm to reduce the dimension of feature points and feature-vectors which are generated by the SIFT algorithm.Solving many defects of the traditional SIFT algorithm like the number of feature points are too many,the dimension of the feature-vector is too high etc. Then we cluster the similar feature points using the cohesive hierarchical clustering algorithm to find out the copy-move forgery area. At the end of the article, we make some experiments to test our algorithm,using 100 pictures in the Columbia University copy-move forgery image library. Results show that the proposed algorithm can generate fewer feature points and lower dimensions of the feature-vector than traditional SIFT, SURF, and PCA-SIFT algorithm,making the efficiency and the accuracy rate of the image forgery detection greatly improved,regardless of how the forgery regions are stretched or rotated. |
| |
Keywords: | |
|
| 点击此处可从《信号处理》浏览原始摘要信息 |
|
点击此处可从《信号处理》下载免费的PDF全文 |