Inconsistency in contrast enhancement can be used to expose image forgeries.In this work, we describe a new method to estimate contrast enhancement from asingle image. Our method takes advantage of the nature of contrast enhancementas a mapping between pixel values, and the distinct characteristics itintroduces to the image pixel histogram. Our method recovers the original pixelhistogram and the contrast enhancement simultaneously from a single image withan iterative algorithm. Unlike previous methods, our method is robust in thepresence of additive noise perturbations that are used to hide the traces ofcontrast enhancement. Furthermore, we also develop an e effective method to todetect image regions undergone contrast enhancement transformations that aredifferent from the rest of the image, and use this method to detect compositeimages. We perform extensive experimental evaluations to demonstrate theefficacy and efficiency of our method method.
展开▼
