作者: David Garcia , Dimitris Stavrakoudis , Antonio Gonzalez , Raul Perez , John B. Theocharis
DOI: 10.2991/IFSA-EUSFLAT-15.2015.180
关键词:
摘要: The inherent interpretability properties of fuzzy rule-based classification systems (FRBCSs) are undoubtedly one their major advantages when compared to conventional black-box classifiers. In this paper we present a preliminary study how the socalled technique feature construction can prove useful in context land cover tasks using remotely sensed imagery. method is integrated into previously proposed genetic FRBCS (GFRBCS) and applied crop task multispectral satellite image. experimental analysis shows that effectively identify very hidden relationships among initial variables problem.