Sign Language is a complex way of communication in which hands, limbs, head, facial expressions and body language play an important role for understanding between deaf-and-dumb people without the use of sounds. In this paper, we propose two methods for Sign Language Recognition using the SVM classifier and features extracted from Hu and Zernike Moments. In the experiments, a comparison between the proposed methods using a database composed of 2040 images for recognition of 24 symbol classes is performed. The results obtained by the method using the Zernike moments features overcomes the ones obtained by the method using the Hu moments achieving an accuracy rate about 96% which is comparable to the ones found in the literature, which holds that our proposal is promising.