Image signal‐to‐noise ratio estimation using adaptive slope nearest‐neighbourhood model

Abstract

A new technique based on nearest neighbourhood method is proposed. In this paper, considering the noise as Gaussian additive white noise, new technique single-image-based estimator is proposed. The performance of this new technique such as adaptive slope nearest neighbourhood is compared with three of the existing method which are original nearest neighbourhood (simple method), first-order interpolation method and shape-preserving piecewise cubic hermite autoregressive moving average. In a few cases involving images with different brightness and edges, this adaptive slope nearest neighbourhood is found to deliver an optimum solution for signal-to-noise ratio estimation problems. For different values of noise variance, the adaptive slope nearest neighbourhood has highest accuracy and less percentage estimation error. Being more robust with white noise, the new proposed technique estimator has efficiency that is significantly greater than those of the three methods.

Nonlinear least squares regression for single image scanning electron microscope signal‐to‐noise ratio estimation

“…Later, Sim & Teh () developed SNR estimation method using adaptive slope nearest neighbourhood model. This method attempts to reduce the dependency on the nature of the SEM images in estimating the SNR value, by adding slope constants into the SNR estimation formula.…”