Detection and Localization of the Copy-paste in Digital Cartoon

Abstract

The application of digital images is becoming more and more widespread, at the same time, image tampering cannot be ignored. Copy-paste tampering usually refers to the original part of the image copied and pasted to other parts of the image, thus hiding the original information and even directly tampering with the original image. In this paper, focusing on the less researched digital comics, we present a new detection and positioning method on tampering, which improves the efficiency of operation and reduce the missed detection while ensuring the accuracy of detection. This method first blocks the image and then extracts the appropriate eigenvalues from blocks. The experimental results show that the proposed method can effectively detect and locate the copy-paste tamper area in digital comics. The error detection rate is low and the calculation speed is fast.