Comparing the performance of fuzzy and crisp classifiers on remotely sensed images: a case of snow classification

摘要: This study deals with the evaluation of accuracy benefits offered by a fuzzy classifier as compared to hard classifiers using satellite imagery for thematic mapping applications. When crisp approach is adopted classify moderate resolution data, presence mixed coverage pixels implies that final product will have errors, either omission or commission, which are not avoidable and solely due spatial data. Theoretically, soft affected such in principle can produce classification more accurate than any classifier. In this we use Pareto boundary optimal solutions quantitative method compare performance statistical one two classifiers, determine highest could be achieved classifiers. As an application, applied case snow from Moderate-Resolution Imaging Spectroradiometer (MODIS) data on alpine sites, validated contemporaneous fine-resolution Advanced Spaceborne Thermal Emission Reflection Radiometer (ASTER) The results showed only outperformed but also yielded higher maximum theoretical areas. While providing general assessment framework obtained inter-comparison exercise effective solution overcome errors intrinsic coarse