Watermarking-based Image Authentication with Recovery Capability using Halftoning and IWT

In this paper we present a watermarking algorithm for image content authentication with localization and recovery capability of the modified areas. We use a halftone image generated by the Floyd-Steinberg kernel as an approximate version of the host image. We adopt this halftone image as a watermark sequence and embed it using the quantization watermarking method into the sub-band LL of the Integer Wavelet Transform (IWT) of the host image. Due to the fact that the watermark is embedded into the sub-band LL of IWT, the proposed method is robust to JPEG compression. Moreover, we employ a Multilayer Perceptron neural network (MLP) in inverse halftoning process to improve the recovered image quality. Using the extracted halftone image, the gray-scale of the modified area is estimated by the MLP. The experimental results demonstrate the effectiveness of the proposed scheme.