Image forgery detection using Gabor filters and DCT

Determining the authenticity of an image has become crucial due to a widespread use of images in various media to covey real or fake messages. In this paper, an image forgery detection method based on Gabor filters and discrete cosine transform (DCT) is proposed. The output of this method is to determine whether an image is authentic or forged. In this method, first, the input image is converted to gray scale image. Second, several Gabor filters with different scales and orientations are applied to the image. Then the DCT from all the filter outputs (subbands) is calculated. The first N coefficients of the DCT from all the subbands are concatenated to form the feature vector. Some feature selections are used to find optimal feature set. Support vector machine (SVM) is used as a classifier. In the experiments, the proposed method outperforms some state-of-the-art methods of image forgery detection.