A Feature Based Approach to Automatic Change Detection from Lidar Data in Urban Areas

Automatic change detection particularly in urban areas is the important tools for their management and planning. This paper presents a feature based method to detect changes in urban area using two LIDAR data sets acquired at different times. The main processes in the method are to detect the change areas through the subtraction between the two DSMs generated from the two individual LIDAR sets, to organize the LIDAR points within the detected areas into surface patches, to classify each patch to one of the pre-defined classes such as ground, vegetation, or building, and to determine the types of the change based on the classes and properties of the patches. The results from the application of the method to real data were verified with the reference data manually acquired from the visual inspection of the orthoimages in the same area. With the proposed method, we were able to detect not only the change area but also the types of the changes in a sufficient degree of accuracy with a reasonable processing time. * Corresponding author.