关键词: 自适应区域生长, GVF-Snake模型, 高分辨率遥感影像, 道路提取

Abstract: Based on the gray characteristic distribution of the objective to be segmented，an adaptive region growing algorithm is proposed，which can estimate the parameters of homogeneity criterion automatically.And the region growing algorithm with the GVF（Gradient Vector Flow）-snake model is employed to extract roads from high-resolution remote sensing images.In the method，the adaptive region growing algorithm is firstly applied to the preliminary road segmentation，and then mathematical morphology is utilized to eliminate disturbances inside and get the outline of the road in the grown image.Finally，it uses the outline as the initial contour of the GVF-snake model，and applies the model to tracking the road，achieving the final result of the road extraction.Experimental results show that the method is efficient and practical for extracting roads from high-resolution remote sensing images，and has a certain adaptive ability.

Key words: adaptive region growing, GVF-snake model, high-resolution remote sensing images, road extraction