In this paper, several wavelet-based techniques for removal of gaussian noise in astronomical images are compared. Validation experiments were carried out with previously processed images from the SDSS, by adding noise artificially and comparing the effectiveness of each method in the reconstruction of original image by means of quantitative indicators like the mean square error and the signal to noise ratio. Results obtained with some of used methods are quite satisfactory. Also, their comparison with other papers shows contradictory conclutions about the best option among the considered methods. Authors of those papers made their validation experiments with another kind of images of non-astronomical origin, having therefore other characteristics.