다단계 계층군집 영상분류법을 이용한 토지 피복 분석

This study used the multistage hierarchical clustering image classification to analyze thesatellite images for the land-cover types of an area in the Korean peninsula. The multistage algorithmconsists of two stages. The first stage performs region-growing segmentation by employing a hierarchicalclustering procedure with the restriction that pixels in a cluster must be spatially contiguous, and finally thewhole image space is segmented into sub-regions where adjacent regions have different physicalproperties. Without spatial constraints for merging, the second stage clusters the segments resulting fromthe previous stage. The image classification of hierarchical clustering, which merges step-by step twosmall groupds into one large one based on the hierarchical structure of digital imagery, generates ahierarchical tree of the relation between the classified regions. The experimental results show that thehierarchical tree has the detailed information on the hierarchical structure of land-use and more detailedspectral information is required for the correct analysis of land-cover types.