Reconstruction of Three Dimensional City Model Based on LIDAR

Abstract:

The research of Three Dimensional City Model (3DCM) has become a hot topic in GIS field in recent years, and it also has played an important role in traffic, land, mining, surveying and mapping, and other fields, especially in urban planning. However, the difficulty to acquire 3D data is the key obstacle to the further development of 3DCM. Airborne LIDAR, integrating GPS, INS and scanning laser rangefinder, can rapidly acquire the 3D position of ground by airplane, which is very economical, efficient and convenient to acquire 3D data. Because traditional three-dimensional data acquisition method can’t meet the need of the city’s fast development, airborne LIDAR technology is regarded as a convenient, swift, high-efficient three-dimensional data acquisition method. Compared with traditional methods, the airborne LIDAR technology has the following characteristics: 1) High efficiency: in 12 hours, the airborne LIDAR can scan 1000 square kilometers, next, with the help of the related post-processing software, LIDAR cloud data can transform them into GIS format or other receivable format in certain automatic or semiautomatic mode. 2) High precision: because the pulse of laser light isn’t easily subject to shadow and solar angle, it greatly improves the data quality. The flight height limit has no influence on its elevation data precision, which is superior to the conventional photogrammetry. The plane precision may achieve 0.15 to 1 meter, the elevation precision may achieve 10 centimeters. 3) All-weather feature: airborne LIDAR is active remote sensing without considering the digital aerial photogrammetry. 4) Rich information: with the aid of airborne LIDAR ,we can obtains not only the three dimensional coordinate of ground point, but also the three dimensional coordinate of terrain details, such as trees, buildings, roads. If it is integrated with CCD, it could gains image information. We acquired the airborne LIDAR data of 20 square kilometers in the central area of Shanghai using ALTM3100 airborne LIDAR system of the Optech company in 2006.This paper introduces the data processing procedure of the airborne LIDAR data, LIDAR system uses random commercial software to process plane GPS tracking data、plane attitude data、 laser ranging data and the swinging angle data of laser scanning mirror, finally, obtaining the three-dimensional coordinates(X,Y,Z) data of various surveying points. Which three-dimensional discrete dot matrix data is without attribute suspending in the air namely LIDAR original data, named “point cloud”. LIDAR data performs pre-processing to obtain digital surface model (DSM), which is classified and extracted, we acquire topography and object related to modeling, preparing for three-dimensional city model. Data pre-processing includes abnormal point deletion, coordinate transformation and flight strip combination. At present, we used famous business software TerraSolid, developed by Company of Finland to realize the classification and extraction from the LIDAR data TerraSolid depends on MicroStation platform, on the basis of classification and extraction algorithms presented by Axelsson, et al. of Swedish, including a lot of module such as TerraScan, TerraModeler and TerraPhoto. TerraScan is used in the field of LIDAR data classification and extraction, TerraModeler is used for producing and dealing with various planes, TerraPhoto is used for dealing with the primitive image, topography model and building model are got by using this software, complicated artificial building (Oriental Pearl, Jin Mao mansion etc.) need artificial repair and disposal, data processing of 20 sq. km. takes more than one month, efficiency has been improved greatly on the premise of guaranteeing the precision. Topography model and building model can be obtained by using TerraSolid and combining a few manual intervention based on DSM, The topography model is expressed with the triangulated irregular network (TIN), the building model is expressed with 3ds format, three dimensional model of non - texture of Lujiazui region of Shanghai was gained by LIDAR data. In order to achieving better visualization effect, the topography model overlaps orthophoto, and stuck true texture to building model, true city landscape of Lujiazui region of Shanghai is established. This paper has introduce post-processing procedure of airborne LIDAR data systematically, has realized the fast reconstruction of three-dimension urban model based on LIDAR data, enable this technology to serve the information construction of the city better.

• Periodical: Advanced Materials Research (Volumes 518-523)
• Pages: 5648-5655
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.518-523.5648
• Online since: May 2012
• Authors: Hui Lin, Ya Zhou Ji, Liang Liang, Wei Liu, Zhao Ling Hu
• Keywords: 3D Data Acquisition, 3DCM, Airborne Lidar, Data Classification and Extraction, DSM, Point Cloud

References

[1] Baatz, M. and Schape, A. Multiresolution segmentation: an optimization approach for high quality multi-scale image segmentation. Angewandte Geographische Informations verarbeitung, 2-23(2000).

[2] Luo Z.Q., Zhang,H.R, et al. Airborne LIDAR technology. The informationization of National territory resources, 47-52(2006).

[3] Wang, J.Z., Li C. M., Lin Z. J. Discussion on acquiring the three-dimensional city technology. The Science of Surveying and Mapping, pp.71-74(2004).

[4] Li, B.J., Fang, Z.X., Ren, J. The research of extracting the characteristic of buildings from the laser scanning data. The Journal of Wuhan University (the version of information science), 65-69(2003).

[5] Long, H.J., Su, L. The present situation of extracting buildings based on the airborne laser scanning data. The science of Surveying and Mapping, 113-116(2005).

[6] Lu, X.C. The three-dimensional scene simulation based on laser scanning data. Journal of system simulation, 176-179(2006).

[7] Liu, J.N. and Zhang, X.H. The development and present situation of surveying the elevation by the laser scanning. Journal of Wuhan University (the version of information science), 132-137(2003).

[8] Li, Y.C. Wen, W.G. Wang, W. The LIDAR technology system of obtaining the ground hree-dimensional data fast. The science of Surveying and Mapping, 35-39(2002).

[9] Sun, M., Ma, A.N., Chen, J. Comments on the present situation of the three-dimensional city model .Journal of Remote Sensing, 155-159(2002).

[10] Axelsson, P. DEM Generation from Laser Scanner Data using Adaptive TIN Models. IAPRS, 110-117(2000).

[11] Vosselman, G. Slope based filtering of laser altimetry data. IAPRS, 935-942(2001).

[12] Sithole, G., Vosselman, G. Comparison of filtering algorithms. ISPRS Report, 147-152(2003).

[13] Wack, R., Wimmer, A. Digital Terrain Models From Airborne Laser Scanner Data-A Grid Based Approach. ISPRS,293-296(2002).

[14] Rottensteiner, F., Briese, C. A New Method for Building Extraction in Urban Areas from High Resolution LIDAR Data. ISPRS, 458-463(2002).

[15] Vosselman, G., Dijkman, S. 3D Building Model Reconstruction from Point Cloud and Ground Plan. International Archives of Photogrametry and Remote Sensing, 37-43(2001).

[16] Morgan, M., Habib, A. Interpolation of LIDAR data and automatic building extraction. ACSM-ASPRS ANNUAL CONFERENCE PROCEEDINGS, 85-91(2002).

[17] Mass, H. G., Vosselman, G. Two algorithms for extracting buildings models from raw laser altimetry data. ISPRS Journal of Photogrammetry and Remote Sensing, 153-156(1999).

DOI: 10.1016/s0924-2716(99)00004-0

[18] You, H.J., Su, L.et al. The fast acquisition and data processing of the three-dimensional remote sensing information of city. The science of Surveying and Mapping, 26-29(2002).

[19] Alharthy, A., Bethel, J. Heuristic Filtering and 3D Feature Extraction from LIDAR Data. ISPRS, 654-660(2002).