Share Email Print

Proceedings Paper

Land cover/use mapping using multi-band imageries captured by Cropcam Unmanned Aerial Vehicle Autopilot (UAV) over Penang Island, Malaysia
Author(s): Tan Fuyi; Beh Boon Chun; Mohd Zubir Mat Jafri; Lim Hwee San; Khiruddin Abdullah; Norhaslinda Mohammad Tahrin
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

The problem of difficulty in obtaining cloud-free scene at the Equatorial region from satellite platforms can be overcome by using airborne imagery. Airborne digital imagery has proved to be an effective tool for land cover studies. Airborne digital camera imageries were selected in this present study because of the airborne digital image provides higher spatial resolution data for mapping a small study area. The main objective of this study is to classify the RGB bands imageries taken from a low-altitude Cropcam UAV for land cover/use mapping over USM campus, penang Island, Malaysia. A conventional digital camera was used to capture images from an elevation of 320 meter on board on an UAV autopilot. This technique was cheaper and economical compared with other airborne studies. The artificial neural network (NN) and maximum likelihood classifier (MLC) were used to classify the digital imageries captured by using Cropcam UAV over USM campus, Penang Islands, Malaysia. The supervised classifier was chosen based on the highest overall accuracy (<80%) and Kappa statistic (<0.8). The classified land cover map was geometrically corrected to provide a geocoded map. The results produced by this study indicated that land cover features could be clearly identified and classified into a land cover map. This study indicates the use of a conventional digital camera as a sensor on board on an UAV autopilot can provide useful information for planning and development of a small area of coverage.

Paper Details

Date Published: 8 November 2012
PDF: 6 pages
Proc. SPIE 8540, Unmanned/Unattended Sensors and Sensor Networks IX, 85400S (8 November 2012); doi: 10.1117/12.974911
Show Author Affiliations
Tan Fuyi, Univ. Sains Malaysia (Malaysia)
Beh Boon Chun, Univ. Sains Malaysia (Malaysia)
Mohd Zubir Mat Jafri, Univ. Sains Malaysia (Malaysia)
Lim Hwee San, Univ. Sains Malaysia (Malaysia)
Khiruddin Abdullah, Univ. Sains Malaysia (Malaysia)
Norhaslinda Mohammad Tahrin, Univ. Sains Malaysia (Malaysia)

Published in SPIE Proceedings Vol. 8540:
Unmanned/Unattended Sensors and Sensor Networks IX
Edward M. Carapezza; Henry J. White, Editor(s)

© SPIE. Terms of Use
Back to Top