Florence Ramirez, Val Anthony Camposano, Marx Sandino Pelone, Samantha Francine Cañete, Danilo Delizo, Renante Violanda, Roland Emerito Otadoy
Wetlands were mapped by employing the Random Trees (RT) classifier on a high resolution orthorectified digital aerial photograph (orthophoto) and LiDAR data of a 50 sq. km area of Toledo, Cebu, Philippines. The rule sets for wetland classification are developed in the eCognition Developer software. Two cases were investigated, namely; wetland extraction using (a) orthophoto and LiDAR data, and (b) LiDAR data only. The obtained producer accuracy for wetland extraction is 97% for both cases, while 91% and 81% user accuracies were obtained for cases (a) and (b), respectively. The overall Kappa values for classifying wetlands and non-wetlands in cases (a) and (b) are 0.89 and 0.78, correspondingly, indicating a reliable classification process. There are 600 wetlands extracted from (a) and 893 from (b). Only 23.5% (141) of the extracted wetlands in (a) and 17.3% (155) in (b) are actual wetlands, yielding to a percent difference of 9.9%.
Phil-LiDAR Research Center, Fr. Josef Baumgartner Learning Resource Center, University of San Carlos, Talamban, Cebu City, Cebu, 6000, Philippines; Physics Department, Theoretical and Computational Sciences and Engineering Group, University of San Carlos, Talamban, Cebu City Cebu, 6000, Philippines
Research at a Glance
Register to unlock
Topics & SDG Alignment
Register to unlock
Collaboration
Register to unlock
Author Profile (Selected)
Register to unlock
References Overview
Register to unlock
Journal & Source
Register to unlock
Metadata & Integrity
Register to unlock