JPEG compression history detection based on detail deviation

The authenticity of the image is crucial to many cases. The efficient detection of the JPEG compression history of bitmap image could reveal the possibility of tampering on the image. In this paper, we propose a lightweight but reliable JPEG compression detection method based on image information loss. An efficient feature of the decreasing percentage of zero coefficient is proposed to detect the JPEG compression history of an image, due to the increasing JPEG compression quality factor. In our method, estimated original images are first created. Then the given image and its estimated counterpart are compressed to get the JPEG coefficient. After that, the image information loss will be calculated. Through the analysis, the goal of the compression history detection can be achieved. Extensive experimental results have demonstrated that the proposed method outperforms the existing methods.