This tool will perform a Kolmogorov-Smirnov (K-S) test for normality to evaluate whether the frequency distribution of values within a raster image are drawn from a Gaussian (normal) distribution. The user must specify the name of the raster image. The test can be performed optionally on the entire image or on a random sub-sample of pixel values of a user-specified size. In evaluating the significance of the test, it is important to keep in mind that given a sufficiently large sample, extremely small and non-notable differences can be found to be statistically significant. Furthermore statistical significance says nothing about the practical significance of a difference.

See Also

two_sample_ks_test

Function Signature

def ks_normality_test(self, raster: Raster, output_html_file: str, num_samples: int) -> None: ...

Project Links

WbW Homepage User Manual Support WbW