Based On The Typical Urban Classification Based On Vehicle Laser Scanning Data

With the development of geographic information systems industry, the rapid access to three dimensional geo-spatial information became an attractive and emerging technology. The vehicle-based scanning technology can quickly obtain the high accurate three-dimensional earth surface information, and as a new method of data collection acquisition, it has been gradually applied to geographic information system. The classification study work of the laser scanning data is the premise and key to the feature extraction and the model building of surface features. The laser data after the classification can be processed in different way according to the needs of users and application areas of different strategies. As the vehicle-based laser data were composed only by discrete points. It obtain the high-precision information of object and did not have continuity in the spatial space. So many methods for two-dimensional image data or airborne laser scanning are not suit for the processor of vehicle-based laser data. Nowadays, the classification technology of the vehicle-based laser scanning data is not mature. To solve these problem, this paper presents a classification method based on gird and segmentation and mainly fit for the city's typical object. The method consider the car collection, spatial distribution of features and geo-spatial characteristics of various types of feature in city. In this paper, the composition of vehicle systems and data acquisition and pretreatment are introduced firstly. And then it analyzed other scholars of laser scan data classification. On this basis, it proposes a Grid-based segmentation method for the classification of other feature. The method in this paper is mainly for the classification for the laser data of the city. It extracts the laser points of the ground through taking three items into account,that is,the data collection characteristics,the vehicle's GPS trajectory, and the slope relationship between the adjacent laser scanning point in the same scanning line, then classify other feature with the method of grid and segmentation. Finally, this work takes the Sanya city as our study area to verify the validly of this method. The classification result is analysid, then the principle of threshold or parameters selection and the effect to the result of experiment is discussed. Hand-foot tray round is used for extracting building feature line. The focus of the future research is presented finally.