Abstract
Invasive species (also known as exotic species) are prime focus of government departments, plant conservationist, ecologist, environmentalist and farmers now days. One of the major non climatic drivers of global change are these plant invasive species that pose serious damages to anthropogenic activities and biodiversity. The study aims at mapping plant invasive species in Karachi using high resolution satellite imagery. World View (WV2) satellite imagery was used as high resolution satellite data source. The imagery was pan-sharpened to the spatial resolution of 0.5 m. Principal component resolution merging technique was used for pan-sharpening of WV 2 imagery. Three level image segmentation was performed using eCognition software. We performed quadtree at first level then multiresolution segmentation was performed at second and third level with different parameters at each level. Once, we picked the shrub canopies well with image segmentation, the WV 2 image was processed for vegetation indices. Threshold classification was performed for all three vegetation indices: ratio vegetation index (RVI), normalized difference vegetation index (NDVI) and enhanced vegetation index EVI2. We conclude that EVI2 mapped the invasive species with the less generalizations and mix classes as compared to RVI and NDVI.
