Image splicing detection based on convolutional neural network with weight combination strategy

Abstract

With the rapid development of splicing manipulation, more and more negative effects have been brought. Therefore, the demand for image splicing detection algorithms is growing dramatically. In this paper, a new image splicing detection method is proposed which is based on convolutional neural network (CNN) with weight combination strategy. In the proposed method, three types of features are selected to distinguish splicing manipulation including YCbCr features, edge features and photo response non-uniformity (PRNU) features, which are combined according to weight by the combination strategy. Different from the other methods, these weight parameters are automatically adjusted during the CNN training process, until the best ratio is obtained. Experiments show that the proposed method has higher accuracy than the other methods using CNN, and the depth of the CNN in the method proposed is much less than the compared methods.