Digital watermarking, as an efficient supplemental method of traditional cryptographic system, has been an important technique for intellectual property protection of digital multimedia. Nowadays, there is an unprecedented development in the image watermarking field. On the other hand, attacks against watermarking systems have become more sophisticated. In general, these attacks can be categorized into common signal processing and geometric distortion. Geometric distortion is known as one of the most difficult attacks to resist. Geometric distortion desynchronizes the location of the watermark and hence causes incorrect watermark detection. Based on HarrisLaplace theory and pseudoZernike moments, a new featurebased image watermarking scheme robust to geometric attacks is proposed in this paper. Firstly, the HarrisLaplace detector is utilized to extract steady feature points from the host image; then, the local feature regions (LFR) are ascertained adaptively according to the feature scale theory, and they are scaled to a standard size; finally, the digital watermark are embedded into the local feature regions (LFR) by quantizing the magnitudes of the pseudoZernike moments. Experimental results show that the proposed scheme is not only invisible and robust against common signals processing such as median filtering, sharpening, noise adding, JPEG compression, etc., but also robust against the geometric attacks such as rotation, translation, scaling, row or column removal, shearing, local geometric distortion, combination attacks, etc.