各种数字图像编辑软件的普及让人们的生活更加多姿多彩,但同时也带来了更多的安全隐患——恶意的篡改者伪造图像、传播虚假内容的门槛极大地降低了。图像局部拷贝—移动(copy-move)是最为容易操作且有效的图像篡改手段之一,针对这种图...各种数字图像编辑软件的普及让人们的生活更加多姿多彩,但同时也带来了更多的安全隐患——恶意的篡改者伪造图像、传播虚假内容的门槛极大地降低了。图像局部拷贝—移动(copy-move)是最为容易操作且有效的图像篡改手段之一,针对这种图像篡改手段的取证技术也在近年来受到了广泛关注。在图像的局部拷贝—移动篡改过程中,通常会包含一定程度的旋转操作,因此取证手段对于旋转的鲁棒性非常重要。本文提出了以筛选后的哈里斯角点结合MROGH(Multi-support Region Order-based Gradient Histogram)特征描述子的图像局部拷贝—旋转—移动篡改取证方法。实验证明,该方法具有良好的特征点覆盖率和旋转鲁棒性。展开更多
Social network platforms such as Twitter, Instagram and Facebook are one of the fastest and most convenient means for sharing digital images. Digital images are generally accepted as credible news but, it may undergo ...Social network platforms such as Twitter, Instagram and Facebook are one of the fastest and most convenient means for sharing digital images. Digital images are generally accepted as credible news but, it may undergo some manipulations before being shared without leaving any obvious traces of tampering; due to existence of the powerful image editing softwares. Copy-move forgery technique is a very simple and common type of image forgery, where a part of the image is copied and then pasted in the same image to replicate or hide some parts from the image. In this paper, we proposed a copy-scale-move forgery detection method based on Scale Invariant Feature Operator (SFOP) detector. The keypoints are then described using MROGH descriptor. Experimental results show that the proposed method is able to locate and detect the forgery even if under some geometric transformations such as scaling.展开更多
提出了一种结合颜色不变量和MROGH(multi-support region order-based gradient histogram)的特殊纹理影像匹配方法。首先在MROGH算法基础上引入颜色不变量模型,构造同时纳入局部影像颜色和空间关系的描述符;然后采用预测同名区域方式...提出了一种结合颜色不变量和MROGH(multi-support region order-based gradient histogram)的特殊纹理影像匹配方法。首先在MROGH算法基础上引入颜色不变量模型,构造同时纳入局部影像颜色和空间关系的描述符;然后采用预测同名区域方式缩小搜索空间;最后采用最小二乘匹配方法精化匹配结果。实验表明,在纹理重复和纹理贫乏区域本文提出影像匹配算法不仅能获取较多匹配点,而且能达到子像素级的影像匹配精度。展开更多
文摘各种数字图像编辑软件的普及让人们的生活更加多姿多彩,但同时也带来了更多的安全隐患——恶意的篡改者伪造图像、传播虚假内容的门槛极大地降低了。图像局部拷贝—移动(copy-move)是最为容易操作且有效的图像篡改手段之一,针对这种图像篡改手段的取证技术也在近年来受到了广泛关注。在图像的局部拷贝—移动篡改过程中,通常会包含一定程度的旋转操作,因此取证手段对于旋转的鲁棒性非常重要。本文提出了以筛选后的哈里斯角点结合MROGH(Multi-support Region Order-based Gradient Histogram)特征描述子的图像局部拷贝—旋转—移动篡改取证方法。实验证明,该方法具有良好的特征点覆盖率和旋转鲁棒性。
基金The authors would like to thank all anonymous reviewers for their insightful comments. Additionally, This work is supported by the National Natural Science Foundation of China (Grant Number: 61471141, 61301099, 61361166006), the Fundamental Research Funds for the Central Universities (Grant Number: HIT. KISTP. 201416, HIT. KISTP. 201414).
文摘Social network platforms such as Twitter, Instagram and Facebook are one of the fastest and most convenient means for sharing digital images. Digital images are generally accepted as credible news but, it may undergo some manipulations before being shared without leaving any obvious traces of tampering; due to existence of the powerful image editing softwares. Copy-move forgery technique is a very simple and common type of image forgery, where a part of the image is copied and then pasted in the same image to replicate or hide some parts from the image. In this paper, we proposed a copy-scale-move forgery detection method based on Scale Invariant Feature Operator (SFOP) detector. The keypoints are then described using MROGH descriptor. Experimental results show that the proposed method is able to locate and detect the forgery even if under some geometric transformations such as scaling.
文摘提出了一种结合颜色不变量和MROGH(multi-support region order-based gradient histogram)的特殊纹理影像匹配方法。首先在MROGH算法基础上引入颜色不变量模型,构造同时纳入局部影像颜色和空间关系的描述符;然后采用预测同名区域方式缩小搜索空间;最后采用最小二乘匹配方法精化匹配结果。实验表明,在纹理重复和纹理贫乏区域本文提出影像匹配算法不仅能获取较多匹配点,而且能达到子像素级的影像匹配精度。
