Robust image watermarking using a chirp detection-based technique

A new robust image watermarking algorithm that embeds multiple watermark bits in the host image is proposed. Linear chirps are embedded as watermark messages, where the slopes of the chirp on the time-frequency (TF) plane represent watermark messages, such that each slope corresponds to a unique message. The Hough-Radon transform (HRT) is a widely used tool to detect directional elements that satisfy a parametric constraint in images. Since linear chirps are localised as a straight line in the TF plane, the HRT is used to detect the watermark messages at the receiver. The HRT acts as an error correcting scheme and robustly estimates the slope of the chirps in the presence of any attacks. It is found that the HRT based detector is able to detect the watermark message accurately when the bit error rate is less than 20%. The robustness of the proposed scheme has been evaluated using the checkmark benchmark attacks.