Content-based image authentication framework with semi-fragile hybrid watermark scheme

Abstract

This paper presents a novel content-based image authentication framework which embeds the semi-fragile image features into the host image based on wavelet transform. In particular, a hybrid method which combines DCT (Discrete Cosine Transform) and DWT (Discrete Wavelet Transform) is introduced to improve the semi-fragile characteristic of the watermarking scheme. Zernike moments of the original image are extracted as feature set and the quantized feature set is embedded in the original image. In the watermark retrieval process, content modification is identified by comparing the Euclidean distance of generated and extracted watermark of the tampered image with predefined threshold. The experimental results demonstrate that the newly proposed DWT-DCT hybrid watermark embedding algorithm is robust for JPEG compression and Gaussian noise up to certain limit. © 2011 Springer-Verlag Berlin Heidelberg.