The Research Of Passive Forensics Against Object Removal By Exemplar-based Image Inpainting

The prevalence of image acquisition devices and powerful processing software have given rise to large amounts of doctored images with no obvious traces,triggering a great demand for automatic forgery detection algorithms that can identify the trustworthiness of a candidate image.Especially,passive image forensics has attracted great research interests since it does not require any auxiliary data such as watermarks or signatures and directly identify the authority of images based on inhere statistical properties.For digital images,the manipulations of image objects including object adding,removing or location modifying are of the most attention because these changes may directly mislead the understanding and awareness of the image content.Exemplar-based inpainting used to be a normal mean to restore old images and videos.However,it can also be a useful tool for object removal.Therefore,passive forensics against exemplar-based inpainting confronts both technical challenges and great application potential.This paper focused on forgery detection against object removal by exemplar-based inpainting,and the innovations and contributions are as follows:In order to reduce computational complexity and improve detection precision,this paper presents an efficient forgery detection algorithm for object removal by exemplar-based inpainting,which integrates central pixel mapping(CPM),greatest zero-connectivity component labeling(GZCL)and fragment splicing detection(FSD).CPM speeds up suspicious block search by efficiently matching those blocks with similar hash values and then finding the sus picious pairs.To improve the detection precision,GZCL is used to mark the tampered pixels in suspected block pairs.FSD is adopted to distinguish and locate tampered regions from its best-match regions.Experimental results show that the proposed algorit hm can reduce up to 90% of the processing time and maintain a detection precision above 85%.To enhance robustness to post-processing,this paper models image manipulations as steganography problems,and proposes a novel detection algorithm for object removal by exemplar-based inpainting.It is based on the analysis that object removal by exemplar-based inpainting has to modify many image pixels without considering some inherent properties within the original image,which is similar to what in steganography.Therefore,we exploit multi-scale subtractive DCT coefficients joint probabilities to extract image features and adopt ensemble classifier to identify the trustworthiness of candidate images.Experimental results show that the proposed algorithm achieve high detection accuracy and is robust to JPEG compression and adding Gaussian noise attacks.Up to present,passive forensics against object removal by exemplar-based inpainting is still in its preliminary exploration stage.It is hoped that the present study maybe promote the development of passive image forensics.