An Efficient Way to Detect and Remove Shadows Based on Multiple Light Sources

Abstract

A shadow is shaped when a light is blocked by an opaque object. They are sometimes unwanted as they may cause failure of image analysis and can also cause poor eminence of information. This paper first describes some important techniques that are used for shadow detection such as segmentation, Histogram Matching and some techniques that are used for shadow removal such as Morphological operations. The proposed methodology is to design a system based on three basic aims-the first goal is shadow detection of single and multiple images, second goal is to remove the shadow from the single and multiple images and third is to calculate the different parameter for measuring the quality of shadow removal method. Once the shadows are detected it becomes simple to detect a non shadow area which is estimated using morphological operators but sometimes when the shadow of any image is merged with the foreground object then the detection process becomes more complex.