PROJECT TITLE :

Extraction of Slum Areas From VHR Imagery Using GLCM Variance

ABSTRACT:

Many cities in the worldwide South face the emergence and growth of highly dynamic slum areas, however usually lack detailed info on these developments. On the market statistical data are commonly aggregated to giant, heterogeneous administrative units that are geographically meaningless for informing effective pro-poor policies. General base data neither allows spatially disaggregated analysis of deprived areas nor monitoring of rapidly changing settlement dynamics, which characterize slums. This paper explores the utility of the gray-level co-occurrence matrix (GLCM) variance to distinguish between slums and formal designed-up (formal) areas in very high spatial and spectral resolution satellite imagery like WorldView-two, OrbView, Quickbird, and Resourcesat. Three geographically completely different cities are selected for this investigation: Mumbai and Ahmedabad, India and Kigali, Rwanda. The exploration of the utility and transferability of the GLCM shows that the variance of the GLCM combined with the normalized distinction vegetation index (NDVI) is ready to separate slums and formal areas. The general accuracy achieved is eighty fourp.c in Kigali, eighty seven% in Mumbai, and 88p.c in Ahmedabad. Furthermore, combining spectral data with the GLCM variance inside a random forest classifier leads to a pixel-based mostly classification accuracy of 90percent. The final slum map, aggregated to homogenous urban patches (HUPs), shows an accuracy of 88percent–ninety five% for slum locations depending on the scale parameter.