Image Denoising Algorithm Based on Incoherent Dictionary Learning

For the shortcoming of losting partial texture information with image denoising process, the image denoising algorithm based on incoherent dictionary learning is proposed. Firstly, the noised image is divided into lots of overlapped image patches, and many patches are extracted for dictionary learning. Then we propose incoherent dictionary learning technology, by which incoherent redundant dictionary is obtained. Finally, sparse representation problem is solved to obtain sparse representation coefficients by sparse coding algorithm, and image is restored by these coefficients. Compared with state-of-the-art methods, the PSNR of the proposed algorithm is better, while image detail and texture information can be preserved to improve visual quality. Our experimental results validate the effectiveness of the proposed algorithm.