Phase based forgery detection of JPEG anti forensics

摘要

Many recent techniques in forgery detection tried to counter noise dithering, which is well adopted in removing footprints of JPEG editing in any image tampering (anti forensics) process. In this paper we present a novel idea of detecting this noise dithering based on detecting spatial frequency phase variations for different blocks of suspect-able images. We try to measure the level of coherence of phase values for spatial frequencies of image pixels and detect threshold variations/distributions that would indicate some editing or tampering of images. The proposed technique is robust against noise dithering due to the fact that local homogeneous regions inherit distinctive phase values. These phase values are inconsistent with embedded or dithered noise signals that are considered to be out of phase and can be easily detected in noisy environments. Our proposed technique is compared with recent successful literature techniques for performance with degraded tampered/hacked images. Unlike other literature techniques, Our proposed approach does not require the knowledge of the quantization matrix that is typically adopted in original compression for forgery detection, as will be illustrated.