基于多特征融合的高分辨率遥感影像道路提取方法研究

Research on Road Extraction Method from High Resolution Remote Sensing Images Based on Multi Feature Fusion

为了提高基于高分辨率遥感影像的道路提取精度,本文提出了一种新的组合道路提取算法。该算法的流程为:首先,通过色彩变换(HIS)获取光谱饱和度(Saturation,S)分量;其次,使用最小二乘支持向量机分类器对经多尺度分割得到道路影像对象的多种特征进行模型训练;最后,将训练后生成的分类模型应用于影像道路信息提取中,并使用数学形态法连接因空洞导致提取不连续的道路。使用高分辨率遥感影像实验数据对本文提出的方法进行检验,结果表明,本文提出方法可准确提取道路信息,精度较高,验证了本文方法的可靠性。

In order to improve the accuracy of road extraction based on high-resolution remote sensing images,a new combined road extraction algorithm is proposed in this paper.The flow of the algorithm is as follows:firstly,the spectral saturation(S)component is obtained by color transformation(HIS);Secondly,the least squares support vector machine classifier is used to train the multi-scale features of road image objects;Finally,the classification model generated after training is applied to image road information extraction,and the mathematical morphology method is used to connect the discontinuous roads caused by holes.The experimental data of high-resolution remote sensing images are used to test the method proposed in this paper.The results show that the method proposed in this paper can accurately extract road information with high accuracy,which verifies the reliability of this method.