Abstract: In order to implement image splicing blind detection,a new splicing detection scheme is proposed based on the model consisting of features from hidden Markov model and some image quality metrics(IQMs) extracted from the given test image,which are sensitive to spliced image.This model can measure statistical differences between original image and spliced image.Kernel-based support vector machine(SVM) is chosen as a classifier to train and test the given image.Experimental results show that the new splicing detection scheme has some advantages of highaccuracy and widelapplication.