Land Use and Land Cover Mapping Using Fuzzy Logic

摘要: The problem of mixed pixels has resulted in the low accuracy land cover classification urban areas. Conventional methods for multispectral remote sensing imagery, such as parallelepiped, minimum distance between means, and maximum likelihood, only utilize spectral information consequently have limited success classifying images. Fuzzy techniques allow to a probability belonging (membership) more than one class, through this membership, imprecise nature data is better represented. In chapter, fuzzy logic-based hierarchical method that incorporates both spatial then developed. Meanwhile, likelihood classifier (MLC) knowledge-based expert system are also applied provide comparison references. This proved produce substantial increase maps with traditional approach.