基于JPEG压缩的数字化地质资料篡改探测方法及应用

Forgery detection and application of digital geological images based on JPEG compression techniques

根据JPEG压缩原理,引入块伪影栅格、编码块后验概率以及ASAD等多种篡改探测方法,并将其应用于数字化成果地质资料的真伪检测。这些方法对于复制-粘贴、喷绘、填涂等篡改检测均为有效,且可以实现被篡改区域的定位。论文分别对不同的检测算法进行了验证和对比分析。实验结果表明:这些方法可有效应用于数字地质资料的篡改检测,从而进一步服务于地质资料信息集群化产业化工作,对海量数字化成果地质资料版权保护和真实性检测具有极高的理论和实用价值。

Based on the JPEG compression principle, the three forgery detection methods including block artifact grid （BAG）, the block posterior probability map （BPPM） and the averaged sum of absolute difference （ASAD） methods were investigated, and the methods were applied to detecting the authenticity of the digital geological images. These methods are efficient for the detection of copy-paste, inpainting and filling tampered images, and can be used to locate the tampered regions. The efficiency and the detection results of three different tampered region detecting algorithms were verified and analyzed. The experimental results show that these methods can be effectively used for the digital geologic images tampering detection, and hence further the information service for the clustering and industry of digital geological image were provided. Besides, these methods have high theoretical and practical value for the copyright protection and authenticity detection of massive digital geological image.