Image Forgery Detection Method for Copy- Move and Splicing Attacks Using DCT, DWT And Correlation

Abstract

Images are utilized to improve the news stories in the papers, evidence in the court of law, legal documents and mostly in social networking sites, etc. But with the advancements in image editing tools, images are open to several falsifications; therefore tampered images are easily created that are difficult to be recognized through naked eye. To carry out such forensic analysis, several technological detections have been developed in the literature. However, most of them are less precise and time-consuming. In this paper, the image authentication technique is based on Discrete Cosine Transform, Discrete Wavelet Transform and Correlation which detects the forgery accurately. DCT, DWT are used for dimensionality reduction. The compressed image is partitioned into overlapping blocks of fixed size. Correlation is then performed between the blocks which detects the region of forgery. Edges of forged region are detected by canny edge detector. Proposed method improves the detection time, precision, recall, accuracy. It is robust to rotation, noise, scaling and multiple copy- move forgeries and splicing attacks.