一种改进的SURF古籍图像复制粘贴篡改检测算法

An Algorithm of Improved Copy-paste Tamper Detection forSURF Ancient Images

针对古籍图像复制粘贴篡改检测需求设计了一种算法。该算法以SURF算法为基础,引入小波变换并改进,清除图像中的噪声斑点,提高了Hessian矩阵构建的准确性;引入SLIC(Simple Linear Iterative Clustering)算法并改进,对图像进行分块,减少了图像匹配检索用时;改进欧氏距离匹配阈值选取方式,降低了特征点检测误匹配率;引入图像局部色彩方差,提升图像特征点匹配准确性。结合多种类型的古籍图像篡改实例,并与当前其他主流算法就检测结果进行比较分析。结果表明,该算法在匹配检测准确度和检测用时方面更具优势,更能有效地鉴定出篡改的图像信息。

An algorithm for the detection of copy-paste tampering of ancient images is designed in this paper.Based on SURF algorithm,wavelet transform is introduced in the algorithm to clear the irrelevant spots in the image and improve the accuracy of Hessian matrix.SLIC(Simple Linear Iterative Clustering)algorithm is introduced to partition the image into blocks to improve the efficiency of image matching and retrieval.The Euclidean distance matching threshold selection method is improved to reduce the mismatching rate of feature point detection.Local color variance is introduced to improve the accuracy of image feature point matching.Combined with many kinds of ancient image tampering examples,the detection results are compared with other mainstream algorithms.Experimental results show that this algorithm greatly improves the matching detection accuracy and detection time,and identifies image tampering information more effectively.