期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2010
卷号：XXXVIII - Part 5
页码：639-645
出版社：Copernicus Publications
摘要：Even though the terrestrial LiDAR mobile mapping system is fairly new, it has been rapidly ingested by end users of wide spectrum of disciplines. As more and more applications are enjoying the potential of terrestrial LiDAR systems, refining quality and a ccuracy of LiDAR data is becoming increasingly critical. While the average geo-referencing accuracy of terrestrial LiDAR systems currently reaches the decimetre level under good GPS conditions, further post geo-referencing enhancement still need to be explored to make up for accuracy deterioration under all conditions. Alternatively, users may seek costly solutions such as building ground control stations to overcome this problem. Furthermore, the radiometric intensity is yet another important aspect that needs to be ad dressed. This paper discusses the feasibility of using the theory of data assimilation in enhancing positioning and radiometric intens ity. Two terrestrial LiDAR datasets of the same area were integrated using assimilation method. The resulted model produced better accuracy as indicated by the comparison to a reference shapefile as well as the standard deviations. A Chi-squared test was also performed where the hypothesis of the estimated value is accepted as the true state at the 95% and 99% confidence level for 3D coordinates whereas it is rejected for the intensity
关键词：LiDAR; Terrestrial MMS; Geo-referencing; Data assimilation; Chi-squared
