Exposing copy-move image tampering using forensic method based on SURF

Digital images are displayed and perceived through a wide variety of media channels: newspapers, magazines, TV, Internet etc. When they are used in a sensitive matter such as evidence in a court of law, one major problem arises: the authenticity of the images. Access to complex image-editing software applications gives way to image tampering that is visually imperceptible. A very common way of image tampering is the copy-move attack, where a portion of the image is copied and pasted onto another region in the same image. This copied portion can potentially be rotated and resized before pasting. In order to expose this type of tampering, a forensic method based on SURF is proposed. This method allows the discovery of copy-move attacks and the regions where it occurred, while maintaining rotation and scale robustness.