In this approach, a rotation, scaling and translation invariant blind second generation watermarking technique is proposed using Tchebichef moments of the original image to estimate the geometric distortion parameters of the corrupted watermarked images. The Tchebichef moments of original image can be used as private key of watermark extraction process. The characteristics of the human visual system are incorporated into the watermark embedding, and the embedding process can be performed in any image domain, including spatial and transform domain. A method in DWT domain is proposed in this paper. Independent Component Analysis is adopted by watermark detector so that the watermark can be correctly extracted but not merely detected. The experimental results show that this method has a good robustness to attacks produced by the popular watermarking test software Stirmark.