Shadow Removal of Individual Tree Crowns in a High Resolution Satellite Imagery

Satellite image processing involves many techniques for enhancement/segmentation of raw (Raster) images acquired from cameras or sensors placed on satellites, space probes and aircrafts. These pictures have numerous applications in regular day to day life. Remote sensing is a process of creating thematic maps as spatial sharing of exacting information and is used to discover and isolate the coherent parts of the surface objects (trees/buildings) and the water bodies. Shadow detection and removal is an main preprocessing for improving act of such (segmentation, thing recognition, view study, tracking, etc) vision tasks. Shadow removal of individual tree top in a coniferous vegetation test zone involves complex procedures. For extracting vegetation zone from a satellite, a pair of spectral indices called Normalized Difference Vegetation Index(NDVI) and Saturation Index(SI) are available. The difficulty of shadows occurring in satellite images need to be addressed while segmenting the tree crowns. In the object slanting shadow detection and deletion or removal method, shadow features are taken into deliberation during image segmentation, and then, according to the statistical features of the images, assumed shadows are extracted. In addition to, some dim objects which could be incorrect for shadows are lined out according to object properties and spatial bond between objects. The retrieved image can be obtained by considering the cross projection advance or approach for shadow free image.