A pipeline algorithm for buildings and structures shapes generation using derived LAS dataset: An efficient alternative to manual digitization from orthophotos in flood hazard feature extraction

 

Bernardino J. Buenaobra1, Aure Flo A. Oraya1, Aries Martin P. Openiano1,
L. Mangle1,  Lora Magnolia F. Cubero1, Kirby Henriksen L. Tan1, Arthur Gil T. Sabandal1, Laarlyn N. Abalos1,

Janice B. Jamora1,2, Ricardo L. Fornis1,2 and Roland Emerito S. Otadoy1,3

1USC Phil-LiDAR Research Center, University of San Carlos

Talamban, Cebu City, Philippines 6000

email: bjbuenaobra@usc.edu.ph

2School of Engineering, Department of Civil Engineering, University of San Carlos
Talamban, Cebu City, Philippines 6000
email: ricfornis@gmail.com

3School of Arts and Sciences, Department of Physics, University of San Carlos
Talamban, Cebu City, Philippines 6000
email: rolandotadoy2012@gmail.com

 

KEY WORDS: Floodplain, LiDAR, Computational, Height break, Workflow

 

 

Conclusion

 

We have shown that a computational approach to the generation of shape files for feature extraction in the floodplains modeling using only LAS laser point files will lead to a better results and in greater detail in the height discrimination of structures from otherwise manual or hand edited counterpart. Moreover, the proposed new workflow and method shows a typical finish time of 4.5 hrs. per 5km x 5km floodplain area could be reduced to around 40 minutes by use of a pipeline algorithm to automatically discriminate buildings and structures.

 

Acknowledgments

 

This paper is an output of “Project 8. LIDAR Data Processing and Validation in the Visayas: Central Visayas (Region 7)” under “Phil-LiDAR 1. Hazard Mapping of the Philippines Using LiDAR – Program B. LIDAR Data Processing and Validation by SUCs and HEIs” headed by Dr. Enrico Paringit. The close collaboration between the USC Phil-LiDAR Research Center, University of San Carlos and the Training Center for Applied Geodesy and Photogrammetry, National Engineering Center, University of the Philippines-Diliman is also hereby acknowledged. We thank the Department of Science and Technology (DOST) for funding support. We also thank the University of San Carlos for moral, logistical, and financial support to the Phil-LiDAR 1 and 2 research programs.

 

References

 

Isenberg, M., LAStools – efficient tools for LiDAR processing. version 150330, from http://lastools.org.
Isenberg, M, 2015. Discriminating Vegetation from Buildings
from: http://rapidlasso.com/2014/10/23/discriminating-vegetation-from-buildings/

Gonzalez, R., Woods R., Eddins S. 2009. Digital Image Processing Using Matlab 2nd Ed. pp.15 Ch.2
National Instruments. 2005. Pattern Matching. IMAQ Vision Manual pp. 12-7 Ch.12
Matlab normxcorr2 function from: http://www.mathworks.com/search/site_search.html?                                                              suggestion=&c[]=entire_site_en&q=normxcorr2