Abstract
Urban green space (UGS) plays indispensable role to maintain the ecological balance of a city. Remote sensing & GIS techniques can play vital role in the accurate estimation of the UGS. This study attempts at a new method of extracting the UGS from high resolution imagery data of Gaofen-2 satellite. The proposed method combines the strength of different transformation techniques i.e. KT transform, principal component analysis and NDVI in order to accurately map UGS. Different combinations were checked to classify the Gaofen-2 satellite imagery of Lanzhou city. The classified image was assessed for classification accuracy and it was observed that the overall accuracy was 89.78% with the kappa coefficient of 0.8125. It was observed that the proposed method can yield relatively high accuracy compared to the individual transformations such as NDVI thresholding.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
