Detecting shadows in a segmented Land Use Land Cover image with LIDAR data

Currently, updating Land Used Land Cover map databases requires a costly manual process. One of the main obstacles in trying to automate this process is the loss of spectral information in shadow areas of the aerial or high resolution satellite images. In this paper, we present a method to automate the detection of shadows in aerial images by using an nDSM built with LIDAR technology and metadata from space images. Additionally, we show that it is possible to recover the class information for these shadow regions by including this process of shadow detection in an object-oriented perspective.