On the Fusion of Lidar and Aerial Color Imagery to Detect Urban Vegetation and Buildings

Abstract

Three-dimensional (3D) data from light detection and ranging (lidar) sensor have proven advantageous in the remote sensing domain for characterization of object structure and dimensions. Fusion-based approaches of lidar and aerial imagery also becoming popular. In this study, aerial color (<small>RGB</small>) imagery, along with co-registered airborne discrete lidar data were used to separate vegetation and buildings from other urban classes/cover-types, as a precursory step towards the assessment of urban forest biomass. Both spectral and structural features such as object height, distribution of surface normals from the lidar, and a novel vegetation metric derived from combined lidar and <small>RGB</small> imagery, referred to as the lidar-infused vegetation index (<small>LDVI</small>) were used in this classification method. The proposed algorithm was tested on different cityscape regions to verify its robustness. Results showed a good separation of buildings and vegetation from other urban classes with on average an overall classification accuracy of 92 percent, with a kappa statistic of 0.85. These results bode well for the operational fusion of lidar and <small>RGB</small> imagery, often flown on the same platform, towards improved characterization of the urban forest and built environments.

Remote sensing for assessing vegetated roofs with a new replicable method in Paris, France

“…The accuracy level we reached with our method is lower overall than with LiDAR. In fact, Bandyopadhyay et al 20 obtained an about 92% accuracy in urban vegetation and buildings detection rates. However, the data used for our modeling are much more available than the LiDAR data.…”

“…[16][17][18] The LiDAR can also be used for mapping, for instance to distinguish vegetated areas. 19 It is sometimes mixed with aerial images, as do Bandyopadhyay et al, 20 to distinguish trees from buildings. 21 According to the literature, LiDAR has a better accuracy to differentiate buildings, roofs, and vegetation.…”

Automatic Registration of Optical Images with Airborne LiDAR Point Cloud in Urban Scenes Based on Line-Point Similarity Invariant and Extended Collinearity Equations

“…Additionally, a laser pulse can penetrate the gaps of plant foliage and hit the ground; hence, not only providing an efficient way for high accuracy Digital Elevation Model (DEM) acquisition in regions covered with vegetation, but also being an indispensable means for forestry parameters retrieval. Both spatial and spectral information can be acquired when optical images and LiDAR point clouds are combined, which effectively compensates for the deficiency caused by a single data source; this combination has many potential applications in natural hazard assessment [ 9 , 10 , 11 ], true orthophoto production [ 12 , 13 ], change detection [ 14 , 15 , 16 ], and automatic manmade objects extraction and modeling [ 17 , 18 , 19 , 20 , 21 ], to mention only a few.…”