Image Denoising using Wavelet Cycle Spinning and Non-local Means Filter

Removing as much noise possible in an image while preserving its fine details is a complex and challenging task. We propose wavelet-based non-local means (NLM) denoising method to overcome the problem. Two well-known wavelets: dual-tree wavelet transform (DT-CWT) discrete (DWT), have been used change into several coefficients sequentially. NLM filtering universal hard thresholding with cycle spinning for on approximation detail coefficients, respectively. The inverse two-dimensional DWT was applied modified obtain denoised image. conducted experiments twelve test images set12 data set, adding additive Gaussian white variances of 10 90 increments 10. Three evaluation metrics, such peak signal rate (PSNR), structural similarity index metric (SSIM), mean square error (MSE), evaluate effectiveness proposed method. From these measurement results, outperforms DT-CWT, DWT, almost all levels except level At that level, lower than but better DT-CWT DWT.