To correctly perform a classification accuracy (or error) assessment, it is necessary to systematically compare two sources of information: (1) pixels or polygons in a remote sensing-derived classification map, and (2) ground reference test information (which may in fact contain error). The relationship between these two sets of information is commonly summarized in an error matrix (sometimes referred to as contingency table or confusion matrix). Indeed, the error matrix provides the basis on which to both describe classification accuracy and characterize errors, which may help refine the classification or estimates derived from it.
Compare different error metrics that are based on the error matrix
New