An object-based two-stage method for a detailed classification of urban landscape components by integrating airborne LiDAR and color infrared image data: A case study of downtown Houston

By exploiting high resolution airborne LiDAR data along with color infrared aerial photographs, this research aims to quantify the urban landscape components using an object-based two-stage method in the case of downtown Houston, Texas, USA. The urban landscape components will be identified and classified by integrating spectral information from color infrared aerial photograph and surface geometric information from airborne LiDAR data. In first stage, the color near-infrared aerial photographs are used to segment the scene into image objects. Then, these objects are classified into three broad categories - impervious surface, vegetation, and water surface, based on their spectral and two-dimensional spatial attributes. In the second stage, the normalized Digital Surface Model derived from airborne LiDAR data is introduced into analysis. Two indicators, relative height and roughness, of each vegetation object from the first stage are calculated, and the threshold values are determined to separate vegetation into lawns, shrubs/hedges, and trees. Next, a series of image processing steps are applied to the nDSM to further classify the impervious surface objects into skyscrapers, high-rise buildings, ordinary buildings, streets, highways, and open spaces. The overall classification accuracy is evaluated as high as 94.10%, and the Kappa coefficient is 92.91%. This research suggests that the combination of morphological information of LiDAR data and spectral information from image data renders a powerful tool for a detailed investigation of urban landscape structure.