Image-based method for noise estimation in remotely sensed data - art. no. 67480L

This paper describes the application of the geostatistical method to quantify noise from a compact airborne spectrographic imager (CASI) data set. Estimation of noise contained within a remote sensing image is essential in order to quantify the effects of noise contamination. Noise was estimated from CASI imagery by calculating the noise as the square root of the nugget variance, a parameter of a fitted semivariogram model. The signal-to-noise ratio (SNR) can then be estimated by dividing the mean value by the square root of the nugget variance. Three wavebands 0.46 - 049 mu m (blue), 0.63 - 0.64 mu m (red) and 0.70 - 0.71 mu m (near-infrared) were used in the analysis. A total of five land covers were selected, each representing a common land cover type in the area which are i) bracken ii) conifer woodland iii) grassland iv) heathland and v) deciduous woodland. The results shows that the noise varies in different land cover types and wavelength.