Estimation of green space ratio for assessing urban landscapes using digital surface models and images

This paper proposes a method for estimating the green space ratio in urban areas by using airborne LiDAR and aerial photographs. The index is defined as the ratio of an area occluded by vegetation to the whole of an area in an azimuth-elevation angle space. Vegetation is detected by a combination of segmented LiDAR data point clouds and image brightness data. The occlusion by vegetation is calculated at an arbitrary location. After an occlusion map in the azimuth-elevation angle space is generated, the green space ratio is obtained by calculating the ratio of the area occluded by vegetation to the entire space. The index can be applied to assess local landscapes, and it is expected that it will provide wide-area estimates at low cost. The authors tested the method using airborne LiDAR data measured in last-pulse mode and aerial photographs. The estimated index map was validated with ground truth data, and the error was acceptably low at approximately 4%. We find that the proposed method is effective for estimating the index over a wide area at low cost.