Output details
11 - Computer Science and Informatics
University of Surrey
Image Forensics Using Generalised Benford's Law for Improving Image Authentication Detection Rates in Semi-Fragile Watermarking
<19>The important contribution of this research is its implementation of the Generalised Benford’s Law (GBL) for detection of unknown JPEG compression for semi-fragile image watermarking. GBL has been used since the 1930s for accounting fraud detection but has only recently found widespread applications in data and image forensics as it can detect unnatural and manipulated data with great accuracy. This publication is amongst the first on the application of GBL to image watermarking and forensics. Here, the compression quality factor was estimated by GBL resulting in improved overall accuracy rates in authenticating the tampered regions for semi-fragile watermarking.