基于图像锐化的SIFT图像复制粘贴篡改检测

SIFT Image copy-paste tamper detection based on image sharpening

SIFT算法提取图像的特征点多为图像目标的边缘点,对于较模糊图像而言,直接使用SIFT算法提取特征点后进行特征点匹配,会降低特征点的匹配数量和匹配正确率,甚至对于复制粘贴篡改图像不能做出正确的判断。所以,提出先对被检测图像利用梯度法进行锐化,使图像目标边缘变得清晰,再用SIFT算法提取特征点及后续操作,以提高匹配的正确率。实验表明,改进算法性能有明显提升。

SIFT algorithm is used to extract the feature points of the image,which are mostly the edge points of the image target.For blurred images,the feature points matching can be reduced by using the SIFT algorithm to extract the feature points directly,and the matching number and matching accuracy are reduced.Even copy and paste tampered images can＇t be correctly judged.Therefore,we first sharpen the detected image by gradient method,then make the edge of the image become clearer,then extract the feature points and follow up operations with SIFT algorithm to improve the matching accuracy.Experiments show that the performance of the improved algorithm is obviously improved.