Building Detection in Complex Scenes Thorough Effective Separation of Buildings from Trees

Effective separation of buildings from trees is a major in image-based automatic building detection. This presents a three-step method for effective separation buildings from trees using aerial imagery and lidar data. , it uses cues such as height to remove objects of low such as bushes, and width to exclude trees with small coverage. The height threshold is also used to a ground mask where buildings are found to be separable than in so-called normalized DSM. Second, entropy and color information are jointly applied to easily distinguishable trees. Finally, an innovative-based procedure is employed using the edge orientation from the imagery to eliminate false positive. The improved building detection algorithm has tested on different test areas and it is shown that the offers high building detection rate in complex which are hilly and densely vegetated.