In recent years a number of new methods have been developed to detect image forgery. Most forensic techniques use footprints left on images to predict the history of the images. The images, however, sometimes could have gone through a series of processing and modification through their lifetime. It is therefore difficult to detect image tampering as the footprints could be distorted or removed over a complex chain of operations. In this research we propose digital forensic techniques that allow us to reverse engineer and determine history of images that have gone through chains of image acquisition and reproduction. This thesis presents two different approaches to address the problem. In the first part we propose a novel theoretical framework for the reverse engineering of signal acquisition chains. Based on a simplified chain model, we describe how signals have gone in the chains at different stages using the theory of sampling signals with finite rate of innovation. Under particular conditions, our technique allows to detect whether a given signal has been reacquired through the chain. It also makes possible to predict corresponding important parameters of the chain using acquisition-reconstruction artefacts left on the signal. The second part of the thesis presents our new algorithm for image recapture detection based on edge blurriness. Two overcomplete dictionaries are trained using the K-SVD approach to learn distinctive blurring patterns from sets of single captured and recaptured images. An SVM classifier is then built using dictionary approximation errors and the mean edge spread width from the training images. The algorithm, which requires no user intervention, was tested on a database that included more than 2500 high quality recaptured images. Our results show that our method achieves a performance rate that exceeds 99% for recaptured images and 94% for single captured images.