关键词: 关键词: 高分辨率遥感影像, 势直方图函数, 邻域总变分, 形态学, 建筑物提取

Abstract: Abstract: Concerning the problem of the low accuracy and high requirements for data in the buildings identification and extraction of the high-resolution remotely sensed imagery, a method based on Neighborhood Total Variation (NTV) and Potential Histogram Function (PHF) was proposed. Firstly, the value of weighted NTV likelihood function for each pixel was calculated, and the segmentation were done with region growing method, candidate buildings were selected from the segmentation result with the constraint of aspect ratio and rectangularity. Then, the shadows were detected automatically. At last, shadows were processed with morphology operations,buildings were extracted by computing the adjacency relationship of the processed shadows and candidate buildings, and the building boundaries were fitted with the minimum enclosing rectangle. For verifying the validity of the proposed method, nine representative sub-images were chosen from PLEIADES images covering Shenzhen, China. The experimental results show that the average precision and recall of the proposed method are 97.71% and 84.21% for the object-based evaluation, and it has more 8% increase in the overall performance comparing with two other similar methods.

Key words: Keywords: high-resolution remotely sensed imagery, potential histogram function, neighborhood total variation, morphology, building extraction