A rapid medical image noise variance estimation method

Abstract

Noise in medical images may affect the result of clinical diagnosis. We propose a rapid noise variances estimation method named Gabor Wavelet Laplacian convolution (GWLC). This method allows rapid estimation of image noise variance without complex calculation. We exclude unwanted image edge lines through Gabor wavelet transform edge detection. Henceforth, we can estimate the noise variance of the image through GWLC. A performance comparison between our proposed method and referred method are carried out. GWLC outperforms other related methods.