16 November 2010 Land cover mapping based on a frequency based contextual classifier from remote sensing data over Penang Island, Malaysia
Author Affiliations +
Abstract
Remote sensing data have been widely used for land cover mapping using supervised and unsupervised methods. The produced land cover maps are useful for various applications. This paper presents a technique for land use/cover mapping using THEOS data of the Penang Island, Malaysia. The objective is to assess the capability of a THEOS image to provide useful remotely sensed images for land cover mapping. The land cover information was extracted from the visible digital spectral bands using PCI Geomatica 10.3 software package. A frequency based contextual classifier was applied to the imagery to extract the spectral information from the acquired scene. Contextual classification is employed when neighbouring pixels are taken into account. The accuracy of each classification map was assessed using the reference data set consisted of a large number of samples collected per category. The study revealed that the frequency based contextual classifier produced superior result and achieved a high degree of accuracy. The preliminary result indicates that THEOS image can be provided useful data for remote sensing to retrieve land cover information at local scale.
© (2010) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
H. S. Lim, M. Z. MatJafri, K. Abdullah, "Land cover mapping based on a frequency based contextual classifier from remote sensing data over Penang Island, Malaysia", Proc. SPIE 7857, Multispectral, Hyperspectral, and Ultraspectral Remote Sensing Technology, Techniques, and Applications III, 78571F (16 November 2010); doi: 10.1117/12.869502; https://doi.org/10.1117/12.869502
PROCEEDINGS
8 PAGES


SHARE
Back to Top