Abstract. Attention to tampering by median filtering (MF) has recently increased in digital image forensics. For the MF detection (MFD), this paper presents a feature vector that is extracted from two kinds of variations between the neighboring line pairs: the row and column directions. Of these variations in the proposed method, one is defined by a gradient difference of the intensity values between the neighboring line pairs, and the other is defined by a coefficient difference of the Fourier transform (FT) between the neighboring line pairs. Subsequently, the constructed 19-dimensional feature vector is composed of these two parts. One is the extracted 9-dimensional from the space domain of an image and the other is the 10-dimensional from the frequency domain of an image. The feature vector is trained in a support vector machine classifier for MFD in the altered images. As a result, in the measured performances of the experimental items, the area under the receiver operating characteristic curve (AUC, ROC) by the sensitivity (P TP : the true positive rate) and 1-specificity (P FP : the false-positive rate) are above 0.985 and the classification ratios are also above 0.979. P e (a minimal average decision error) ranges from 0 to 0.024, and P TP at P FP ¼ 0.01 ranges from 0.965 to 0.996. It is confirmed that the grade evaluation of the proposed variation-based MF detection method is rated as "Excellent (A)" by AUC is above 0.9. © The Authors. Published by SPIE under a Creative Commons Attribution 3.0 Unported License. Distribution or reproduction of this work in whole or in part requires full attribution of the original publication, including its DOI.

To attain a robust feature vector for median filtering detection (MFD) in digital forgery images, this paper presents a short feature vector that is made up of three types of feature sets. The first set is defined by the variation to be the 3-D length in the gradient difference of the intensity values of the adjacent row and column line pairs in the image, respectively. The second set is defined by the variation in the coefficient difference of the Fourier transform to be the 3-D length in the adjacent line pairs. The last set is defined by the residual image between an image and its reconstructed image by the gradient based on solving Poisson's equation, which is also the 3-D length. Two of the sets are extracted in the spatial and spectral domains of an image, respectively, and the last set is extracted from the residual image. The totally formed 9-D feature vector is subsequently trained in the support vector machine classifier for MFD. In the experimental results of the proposed variation-and residual-based MFD scheme, the area under the curve is achieved closer to 1. Despite a short feature vector, the evaluation of the proposed MFD scheme is graded as "Excellent (A)". In particular, the scheme detected good median filtering from the JPEG post-compression image for the cut-and-paste forgery image.

To detect median filtering forensics, a four-feature ensemble including the median filter residual autoregressive (MFR AR) model, statistical properties, gradient-edge line, and HU invariant moments of an image were used to propose an improved feature vector. The defined novel feature vector was trained on a support vector machine (SVM) classifier for median filtering detection (MFD) of forgery images. The performance of the proposed MFD scheme was measured with several types of images: median filtered (window size: {3 × 3, 5 × 5, composite (3 × 3, 5 × 5)}), JPEG compressed (quality factor: 90) after median filtered, rotated (counterclockwise: 5 •), and noise added (salt-pepper: 5%) which has been re-altered in various ways. Experimental results show high efficiency and performance of the MFD techniques. The area under the curve (AUC) by sensitivity (TP: true positive rate) and 1-specificity (FP: false positive rate) results of the proposed MFD scheme are 0.9 upper with the trained SVM classifier. Thus, the grade evaluation of the proposed scheme is ''Excellent (A).'' INDEX TERMS Forgery image, median filtering detection, digital image forensics, median filter residual, autoregressive model, HU invariant moments, support vector machine.

Since the ground truth (GT) generated by CNN has pieces of patch information of the learned class, the accurate detection of Copy-Move is ambiguous. By various CNNs for image classification and semantic segmentation, the generated GT images are different yet similar patch patterns for detecting forgery regions, and it is difficult to determine which network model-generated GT image is suitable. Therefore, an optimal GT image is essential in image forensics. The proposed scheme in this paper generates a novelty GT image to solve this problem for the correct detection of Copy-Move forgery. The novelty GT image was configured using image classification and semantic segmentation. The variety of GT images is generated by adopting the state-of-the-art four image classifications and one semantic segmentation in the deep neural network. The proposed scheme implements mainly three tasks: 1) each network model generates the GT images (GTnet), 2) which are convergence synthesized into one (GTconv), and 3) it decomposed again into GT images (GTdecomp) with a threshold value of the 'Threshold Filter.' Here, the GTnet images involve two pieces of information about the image classification and semantic segmentation of the forgery image. The GTconv has two pieces of information as one GT image. The GTdecomp is decomposed GTconv into various GT images by the threshold value, which is a permeated degree of the information about 'Image classification' and 'Semantic segmentation.' With this operational flow, the proposed novelty GT image is accomplished for Copy-Move forgery detection. The results confirmed in the experiment for comparing the performance of the existing GTnet image and the GTdecomp image of the proposed scheme showed that the Accuracy and F1 Score of the proposed scheme had the maximum improvement rate of 0.4% and 0.2%, respectively. Also, by estimating the proposed CMFD scheme, Area Under the Curve (AUC) is graded as 'Excellent (A)' with a value of 0.9 higher.

For the detection of median filtering (MF) forensics, this paper proposes the feature vector extracted from the bit-planes slicing of the forged image. The assembled feature vector is trained in a support vector machine (SVM) classifier for the MF detection (MFD) of the forged images. The performance of the proposed MFD scheme is measured with several types of forged images: unaltered, Gaussian filtering (3×3), averaging filtering (3 × 3), downscaling (0.9), upscaling (1.1), and post-frame-up, respectively, in a block size 32 × 32 and 64 × 64 pixels. Subsequently, in experimental items, a classification ratio, Area Under the Curve (AUC), P TP at P FP = 0.01, and Pe (a minimum average decision error) are estimated. The result in terms of AUC shows that the estimation of the proposed MFD scheme is graded as 'Excellent (A)'. INDEX TERMS Median filtering detection (MFD), forgery image, digital image forensics, bit-planes slicing, residual image, support vector machine (SVM).