| Autor: |
Pan Zhu, Zhong Lu, Xlaoyong Chen, Honda, Klyoshl, Elumnoh, Aplslt |
| Předmět: |
|
| Zdroj: |
Photogrammetric Engineering & Remote Sensing; Dec2004, Vol. 70 Issue 12, p1433-1440, 8p |
| Abstrakt: |
This paper presents an automatic mad extraction technique that combines information from aerial photos and laser scanning data (LSD). This innovative Road Extraction Assisted by Laser (REAL) can detect the mad edges shadowed by surrounding high objects. A new concept of Associated Road Line (ARL) graph from LSD is introduced to enhance Real Road Line (ARL) graph from aerial photos. The extraction process consists of three steps: The first step is analysing laser images where parameters such as height and edges of high objects are obtained. Secondly, digital images are analysed where road edges are detected. It is evident that ARL and RRL graphs are homeomorphous which provides a theoretical foundation of REAL. The gaps of RRL are bridged through their ARL with a topological transformation. Finally, shadowed parts of RRL are reconstructed by a spline-approximation algorithm. The experimental results show that this approach is effective and has potential advantages. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Supplemental Index |
| Externí odkaz: |
|
