Median Filtering Detection of Digital Image using the Pixel’s Gradients
Kang Hyeon Rhee · ITC-CSCC :International Technical Conference on Circuits Systems, Computers and Communications · 2015
For a design of the median filtering (MF) detection in the altered digital images, this paper presents a new feature vector which is formed from AR (Autoregressive) coefficients by AR model of the gradients of the horizontal and vertical lines in an image. In the proposed algorithm, AR coefficients are computed from the gradients of the lines. Subsequently, the defined 10 Dim. feature vector is trained in an SVM (Support Vector Machine) for MF detection in the forged images. On the MF classification, compare to the MFR (Median Filter Residual) method that had the same 10 Dim. feature vectors. In the experiment, three kinds test items are AUC (Area Under Curve), a classification ratio and a minimal average decision error. The performance is excellent at Unaltered, Averaging filtering (3×3) and JPEG (QF=90) images, and less at Gaussian filtering (3×3) image on MF3 detection. On the MF5, the performance of all test items is superior. And on the MF35, the performance is excellent except JPEG. However, in the measured performances of all items, AUC by the sensitivity (TP: True Positive rate) and 1-specificity (FP: False Negative rate) is approached to 1. Thus, it is confirmed that the grade evaluation of the proposed algorithm is ‘Excellent (A)’.