Automatic acquisition of 3D spatial data in city with air-borne TLS Three Line Scanner
wrong correspondence may arise because some buildings may have almost the same characteristics. Another source of wrong correspondence is the occlusion,
i.e. a feature in one image may not appear in the other images. Then, in our study, laser scanner data is used for solving these problems.
Fig5 : Rectified image of TLS
4.3.Ideas of using laser scanner data to support stereo matching process
Laser scanner data can be used for reducing the range of search for corresponding features area of reference, because the laser scanner data give approximate height of the buildings and terrain surfaces.
And before matching, occluded area can be detected from the coarse 3D data from the laser scanner and the estimated position and attitude of the aircraft. The results of the experiment of integrating TLS data and the laser scanner data will be presented at the conference.
5.Summary
We proposed methodological framework of generating three-dimensional data automatically from data of TLS and laser scanner. For the purpose of generating 3D city model with higher resolution and automation level, TLS image can compensate the low resolution of laser scanner data, while the laser scanner data can assist stereo matching process by providing approximate height of terrain surface and building height. Detailed algorithm should be devised using the experimental flight data. The result will be reported at the conference.
References
- Norbert Haala, Dirk
Stallmann and Micheal Cramer, "Calibration of
directly measured. position and attitude by aero
triangulation of three-line-airbone imagery"
- Ryosuke Shibasaki,
1993, "Three Line Scanner", Photogrametry &
Remote Sensing. Vol.32,No.6 pp26-30
- Zhongchao Shi and Ryosuke Shibasaki,1995,"Application of Wavelet Transform Based. Image Segmentation and Stereo Matching for Automated House Detection from Aerial Photograph",Photogrametry & Remoto Sensing Vol.34, No.5
