Wavelet-based watermarking for tamper proofing of still images

We propose a novel robust and fragile watermarking technique for the tamper proofing of still images. In this method, robust watermarks are embedded in the lowest frequency components of an image by using a controlled quantization process, while fragile watermarks, which are used to determine whether the image has been altered, are embedded by thresholding and quantizing wavelet coefficients at the coarser scales. This technique enables us to distinguish malicious changes such as modifying or forging a part of an image from non-malicious changes resulting from common image processing operations like lossy compression. Furthermore, we can detect any change in the image as well as localizing the areas that have been changed.