基于旋转灰度特征与位移约束的图像伪造检测算法

李晓红; 杨玉香; 姜春峰

doi:10.13568/j.cnki.651094.2018.03.011

您当前的位置：

首页 >

文章列表页 >

基于旋转灰度特征与位移约束的图像伪造检测算法

信息科学与技术 | 更新时间：2026-01-21

- 基于旋转灰度特征与位移约束的图像伪造检测算法
- 基于旋转灰度特征与位移约束的图像伪造检测算法
- 新疆大学学报（自然科学版中英文） 2018年35卷第3期页码：314-320
- 作者机构：
  
  1. 吉林省经济管理干部学院
  2. 中国计量学院
  3. 东北大学
- 作者简介：
- 基金信息：
  
  吉林省科学技术厅科技攻关计划重点科技攻关项目(20170204023G X)
- DOI：10.13568/j.cnki.651094.2018.03.011
  中图分类号： TP391.41
- 纸质出版：2018
- 稿件说明：
移动端阅览
[1]李晓红,杨玉香,姜春峰.基于旋转灰度特征与位移约束的图像伪造检测算法[J],2018,35(03):314-320+332.
[1]李晓红,杨玉香,姜春峰.基于旋转灰度特征与位移约束的图像伪造检测算法[J],2018,35(03):314-320+332. DOI： 10.13568/j.cnki.651094.2018.03.011.

DOI：10.13568/j.cnki.651094.2018.03.011.

摘要

当前图像伪造检测方法主要依靠近邻比值法来实现图像内容的真伪决策

由于近邻比值法的判断条件单一

该类算法的检测精度不佳.因此

本文提出了旋转灰度特征耦合位移约束的图像伪造检测算法.首先

根据图像像素点

构造十字约束法则

对FAST算法予以改进

以提取待检测图像中的特征点;然后

利用像素点的梯度特征形成直方图

通过直方图峰值获取特征点的主方向;再用特征点的灰度值来构造旋转灰度特征模型

用于获取特征向量

生成特征描述子;用特征点的位置以及角度特征

构造了位移约束规则

并且在位移约束规则下

通过归一化互相关函数对特征点的相似性进行度量

完成特征点匹配.最后

引入均值漂移模型

对图像中的伪造内容完成区域定位

实现图像的伪造检测.实验结果表明:与当前图像伪造检测算法相比

本文算法不仅具有更高的检测精度以及检测效率

而且还具有更好地鲁棒性能.

Abstract

In view of the current image forgery detection method to achieve image matching and image forgery detection mainly relies on the nearest neighbor ratio method

when the detected image exists in the complex forge content image

this method appears more error detection point and detection of defects such as time-consuming. So an effective image forgery detection algorithm based on rotation gray feature and displacement constraint was proposed in this paper. First of all

the FAST algorithm is used to construct the cross constraint rule

and then the improved algorithm is used to extract the feature points in the image to improve the detection efficiency. Then

the histogram of the pixel is used to form the histogram

and then the main direction of the feature point is obtained by the histogram peak. Then

using the gray value of the feature points to construct the rotation gray feature model

which is used to obtain the feature vector and generate the feature descriptor. Finally

the position of the feature points and feature point

the formation of displacement constraint rules

and displacement constraint rules using normalized cross-correlation function to measure the similarity of feature points

feature point matching. The mean shift model is used to locate the forgery in the image

and the forgery detection is completed. The experimental results show that the forged images were detected when compared with the current image forgery detection algorithm; this algorithm has higher detection precision and efficiency

but also has better robust performance.

关键词

Keywords

references

Zhao Fei,Shi Wenchang,Qin Bo.Image forgery detection using segmentation and swarm intelligent algorithm[J].Wuhan University Journal of Natural Sciences,2017,2(22):141-18.

Emam Mahmoud,Han Qi,Niu Xiamu.PCET based copy-move forgery detection in images under geometric transforms[J].Multimedia Tools and Applications,2016,18(75):11513-11527.

欧红玉,陈曦,宋燕辉.基于LBP的图像复制篡改检测[J].计算机应用与软件,2013,9(30):170-172+178.

李景富,张飞.凸优化耦合传感器模式噪声的图像伪造检测[J].计算机测量与控制,2015,23(5):1678-1681+1685.

Maryam Jaberi,George Bebis,Muhammad Hussain.Accurate and robust localization of duplicated region in copy–move image forgery[J].Machine Vision and Applications,2014,2(25):451-475.

Kalra G S,Talwar R,Sadawarti H.Adaptive digital image watermarking for color images in frequency domain[J].Multimedia Tools and Applications,2014,3(36):416-423.

李晓飞,李鹏飞.基于SIFT的伪造图像盲检测算法[J].长春大学学报,2014,10(24):1354-1357.

柴新新,邱晓晖.基于提升小波变换的图像篡改检测算法[J].计算机技术与发展,2016,4(26):78-81.

王蒙,戴亚平,王庆林.一种新的FAST-Snake目标跟踪方法[J].自动化学报,2014,6(40):1108-1115.

Edward Rosten.Faster and Better:A Machine Learning Approach to Corner Detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2010,1(32):105-119.

王飞宇,邸男,贾平.结合尺度空间FAST角点检测器和SURF描绘器的图像特征[J].液晶与显示,2014,4(29):598-604.

JING Huiyun,HE Xin,HAN Qi.CBRISK:Colored Binary Robust Invariant Scalable Keypoints[J].IEICE TRANSACTIONS on Information and Systems,2013,2(E96):392-295.

Rosten,Drummond T.Machine Learning for High-speed Corner Detection[C].European Conference on Computer Vision,2006:430-443.

朱珏钰.基于点匹配的区域拷贝篡改检测[J].计算技术与自动化,2014,2(33):97-100.

杜振龙,杨凡,李晓丽.利用SIFT特征的非对称匹配图像拼接盲检测[J].中国图象图形学报,2013,4(18):442-449.

董文会,常发亮,李天平.融合颜色直方图及SIFT特征的自适应分块目标跟踪方法[J].电子与信息学报,2013,4(35):770-776.

李慧,蔺启忠,刘庆杰.基于FAST和SURF的遥感图像自动配准方法[J].国土资源遥感,2012,2(6):29-33.

张智丰,裴志利.基于模糊局部二值模式算子的图像伪造检测[J].计算机工程与设计,2015,12(36):3284-3290.

Zheng Jiangbin,Liu Yanan,Ren Jinchang.Fusion of block and keypoints based approaches for effective copy-move image forgery detection[J].Multidimensional Systems and Signal Processing,2016,4(27):989-1005.

焦丽鑫,杜振龙.基于均值漂移的图像复制粘贴伪造盲检测[J].计算机应用,2014,34(3):806-809.

Swapnil H.Kudke,Avinash D.Gawande.Copy-Move Attack Forgery Detection by Using SIFT[J].International Journal of Innovative Technology and Exploring Engineering,2015,2(5):221-224.

赵洁,郭继昌.基于JPEG系数变化率的图像复制粘贴篡改检测[J].浙江大学学报(工学版),2016,10(49):1893-1901.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

暂无数据