Tree Crown Delineation from High Resolution Satellite Images

Abstract

This paper presents the comparison of two different algorithms for detection and counting of individual tree crown from high resolution satellite images. Trees can be detected from the high resolution satellite images using the different algorithms like K-means clustering, Circle Unification, Watershed segmentation and Normalized Difference Vegetation Index (NDVI). In this paper Watershed segmentation algorithm and NDVI performance are improved for the detection and counting the trees in the satellite images. The fundamental process of algorithms and results are compared to find the better approach for the detection of the trees.