Publication Date

Fall 2010

Degree Type


Degree Name

Master of Science (MS)


Electrical Engineering


Essam Marouf


denoising, haar, image, tetrolet, transform, wavelet

Subject Areas

Electrical Engineering; Multimedia; Engineering


A new denoising algorithm based on the Haar wavelet transform is proposed. The methodology is based on an algorithm initially developed for image compression using the Tetrolet transform. The Tetrolet transform is an adaptive Haar wavelet transform whose support is tetrominoes, that is, shapes made by connecting four equal sized squares. The proposed algorithm improves denoising performance measured in peak signal-to-noise ratio (PSNR) by 1-2.5 dB over the Haar wavelet transform for images corrupted by additive white Gaussian noise (AWGN) assuming universal hard thresholding. The algorithm is local and works independently on each 4x4 block of the image. It performs equally well when compared with other published Haar wavelet transform-based methods (achieves up to 2 dB better PSNR). The local nature of the algorithm and the simplicity of the Haar wavelet transform computations make the proposed algorithm well suited for efficient hardware implementation.