This chapter presents a method for detecting and recovering geometrical attacks in digital watermarking by making use of geometric moments of the original images. Digital image watermarking has become a popular technique for authentication and copyright protection. However, many proposed image watermarking techniques are sensitive to geometric distortions, such as rotation, scaling, and translation. In this chapter, we propose a new way of making this estimation by using the geometric moments of original image. The moment information can be used as a private key of extraction process. This method can be used as a preprocess of the extraction watermarking process. We have embedded different watermarks into original images in different domains including discrete wavelet transform (DWT), discrete cosine transfrom (DCT), fast Fourier transform, and spatial domain. The experimental results show that our method has a good robustness to wide geometric distortion parameters ranges and it is robust to Stirmark attacks.