Usage of indices for extraction of built-up areas and vegetation features from landsat TM image: a case of Dar es Salaam and Kisarawe peri-urban areas, Tanzania

Abstract

This paper address the use of Indices Combination with supervision classification methods to extract urban built up areas, vegetation and water features from Landsat Thematic Mapper (TM7) imagery covering Dar es Salaam and Kisarawe peri-urban areas. The study uses three indices; Normalized Difference Built up Index (NDBI), Modified Landsat TM7 image into three thematic-oriented bands. Data correlation, spectral confusion and redundancy between original multispectral bands were significantly reduced upon application of the techniques. As a result, the spectral signatures of the three urban land use classes are well separated. Through a supervised classification on the newly formed image, the urban built-up areas, vegetation and water features were finally extracted effectively; with the accuracy of 82.05 percent attained. The results show that the technique is effective and reliable and can be considered for use in other areas with similar characteristics.