Image denoising by adaptive directional lifting-based discrete wavelet transform and quantization
Naoki Furuhashi, Azusa Oota, Taichi Yoshida, Masaaki Ikehara · 2013
In this paper, we propose the non-local method for image de-noising via adaptive directional lifting-based discrete wavelet transform (ADL) and quantization. The non-local methods such as non-local means are interested in image denoising based on the self-similarity. They search similar blocks and estimate the original value. The proposed method doesn't search but generates new similar blocks by ADL with multiple directions, and quantization to denoise. It improves the denoising quality and reduces the computational complexity. Finally, we compare the proposed and conventional method, and show an advantage of them.