Using internal depth to aid stereoscopic image splicing detection

Abstract

Splicing is a common image manipulation technique, where parts of multiple images are combined to create a new composite image. Commercial image editing software enables almost anyone to splice images and create fake photographs. This paper investigates how the relationship between object distance and internal depth can aid in detecting spliced stereoscopic images. An equation is derived for predicting the distance at which an object loses internal depth. Experiments with stereoscopic images indicate that the analysis of this depth information can assist in detecting image splicing. © 2012 IFIP International Federation for Information Processing.