基于局部特征和投票表决法的居民地提取

Residential Area Extraction Based on Local Feature and Voting Principle

本文针对高分辨率影像中详细的居民地局部信息，提出基于局部特征的居民地提取方法，即以Gabor滤波提取局部特征点，并依据局部特征点建立投票表决模型，然后在选票分布上依据最优阈值分割出居民地区域。实验结果表明：该方法具有良好的居民地分割效果。

As high resolution images could provide detailed local infm^nation of residential area, a method based on local feature was proposed for residential area extraction. Local feature points extracted by Gabor filtering were used to set up a voting model. Then the residential areas were divided by optimal threshold that was determined on the basis of vote distribution. The resuhs of this experiment indicated that the method was effective in dividing residential area.