Detection of manipulated images
When compared to the existing image formats, JPEG is the most widely used format that stores digital images using cameras and software based algorithms. Digital images have widespread use in applications such as surveillance, military, scientific, etc. With the availability of tools on the internet it is easy to manipulate the digital images. For instance, a true image when manipulated becomes useless in certain applications. An image that is supposed to be the source of evidence, if manipulated will be of no use in some applications. Hence there arises a need to verify the originality of digital images without the availability of the original image. In this thesis, we have developed a method to test the originality of a given image. The verification methods are based on computing DCT (Discrete Cosine Transform) for a given image. The pdf plots for the obtained DCT coefficients are plotted to observe the characteristics of the plot with the assumed Laplace distribution function. Later, the parameters of the distribution functions are estimated and computed using Maximum Likelihood Estimation, in order to test whether a given image is natural or manipulated.
Thesis (M.S.)--Wichita State University, College of Engineering, Dept. of Electrical and Computer Engineering