A study of medical image tampering detection

Currently, methods of image tampering detection are divided into two categories, active detection and passive detection. In this paper, we try to review several detecting methods and hope this will offer some help to this field. We will focus on the passive detection method for medical images and show some results of our experiments in which we extract statistical features (IQM and HOWS based) of source images and their doctored version respectively. Manipulations we take to doctor the images include: brightness adjustment, rotation, scale, filtering, compression and so on, using fix manipulation parameter and random selected parameter. Different classifiers are chosen then to discriminate the source images from the doctored ones. We compare the performance of the classifiers to show that the passive detection methods are effective while dealing with medical image tapering detecting.