AUTOMATIC SHADOW DETECTION FOR HIGH-RESOLUTION REMOTE SENSING DATA

Abstract. Shadow can be casted by daylight or any other light sources. We will not get a clear and quality image if it's hovered the shadow. Shadows are frequently formed in high-resolution satellite imagery limitations of imaging environment presence high rise structures, this scenario is true especially metropolitan regions. one noteworthy evils remotely sensed which hinders precision information extraction change identification. To attenuate effects shadow resolution regarding their supplemental functions, our paper suggests novel algorithm for masking built on computational methods. Firstly we transformed images from RGB space to CIELCh model, next evaluated modified Specthem ratio, then used multilevel thresholding. also created masks areas having vegetation, water, soil. mask noise was decreased morphological techniques. The ratio lighting shadowed unshadowed utilized create masks, remove shadows source photos. thresholding approach creates an initial during detection step, filtering method incorrect vectorized raster data further applied various studies.