Comparison of four classification methods to extract land use and land cover from raw satellite images for some remote arid areas, Kingdom Of Saudi Arabia

Abstract EN

Remote sensing (RS) technologies was utilized to extract some of the important spatially variable parameters, such as land cover and land use (LCLU), from satellite images for remote arid areas in Saudi Arabia.

Four different classification techniques unsupervised (ISODATA), and supervised (Maximum likelihood, Mahalanobis Distance, and Minimum Distance) are applied in three sub-catchments in Saudi Arabia for the classification of the raw TM5 images.

The developed maps are then visually compared with each other and accuracy assessments utilizing ground-truths are undertaken.

It was found that the Maximum likelihood method gave the best results and both Minimum distance and Mahalanobis distance methods overestimated agriculture land and suburban areas.

In spite of missing few insignificant features due to the low resolution of the satellite images (90m), good agreement between parameters extracted automatically from the developed maps and field observations was found.