Rod extraction method based on vehicle-mounted laser scanning point cloud data
In response to the problem of unsatisfactory classification accuracy of rods in mobile vehicle-mounted laser scanning point cloud data,this paper constructed a rod classification algorithm based on the support vector machine(SVM)model regarding the spatial morphology characteristics of rods.Firstly,10 feature values were determined,and a feature matrix was established based on the spatial morphology characteristics of the rods.Secondly,the SVM model was trained,and a classification model was established.Finally,the trained optimal SVM model was used for rod classification.A certain section of urban road point cloud data was selected for testing,and the results show that the classification model in this paper does not require manual intervention and threshold setting,and it has a high degree of automation.Its highest classification accuracy of rods can reach 94.23%,verifying its effectiveness and superiority.It can provide some reference for ground object classification based on laser point cloud data.
vehicle-mounted laser scanningpoint cloudrodsupport vector machine(SVM)feature value
国家科技期刊平台
NSTL
万方数据
