Further Study on GFR Features for JPEG Steganalysis

Œe GFR (Gabor Filter Residual) features, built as histograms of quantized residuals obtained with 2D Gabor €lters, can achieve competitive detection performance against adaptive JPEG steganography. In this paper, an improved version of the GFR is proposed. First, a novel histogram merging method is proposed according to the symmetries between di‚erent Gabor €lters, thus making the features more compact and robust. Second, a new weighted histogram method is proposed by considering the position of the residual value in a quantization interval, making the features more sensitive to the slight changes in residual values. Œe experiments are given to demonstrate the e‚ectiveness of our proposed methods. Finally, we design a CNN to duplicate the detector with the improved GFR features and the ensemble classi€er, thus optimizing the design of the €lters used to form residuals in JPEG-phaseaware features.