Combined and optimized 2-steps ratio map for shadow detection in aerial images

Shadow is a malicious phenomenon in aerial image processing algorithms. Locating shadow regions is very helpful in aerial image processing. This paper presents a new optimized algorithm for extracting shadows from a single aerial color image. Most of similar approaches extract shadows by using a suite color space. In this work, two RGB and YCbCr color spaces are combined and a powerful 2-steps ratio map has been created. Furthermore, as a new idea, effect of the blue color of sky is used to improve performance of the proposed ratio map. Candidate shadow and non-shadow regions are separated by applying Otsu's thresholding method. Because of the performance of the ratio map, it is not necessary to analyse the image locally and region by region. Therefore, it leads to decrease computational cost and it can be used in online applications. In order to demonstrate the superior performance of the proposed method, both qualitative and quantitative comparisons of the method with other state-of-the-art techniques are provided.

Automated summary

This paper presents a new optimized algorithm for extracting shadows from aerial images, which combines two color spaces and utilizes the blue color from the sky to enhance the performance of the ratio map. The algorithm reduces computational cost and can be used in online applications.