Advanced Spatial Categorization of Buildings Based on Point based Cloud Data Algorithms

There is a tremendous horizontal and vertical growth, where an immediate demand for geospatial tools for precision urban planning and sustainable development is gaining more interest. Acquisition of high resolution, 3D spatial data through Light Detection and Ranging (LiDAR) technology is an exploitable medium. Traditional grid-based LiDAR methods, however, tend to have data loss and lower accuracy. An automated, point based classification methodology is introduced to further augment the classification of raw LiDAR data for urban areas in Tamil Nadu. Through spatial characteristics of point height, point density and local plane orientation, the proposed method efficiently classifies LiDAR points into ground, vegetation and building classes. By successfully reconstructing 3D urban models, the study was able to reflect large urban clusters in urban centres and sparse low-rise structures in rural areas. These models demonstrate the spatial relations between urban characteristics, they develop urban patterns and fluctuations in eco balances. Results show the capacity of this approach being potentially applicable to urban planning, smart city development, landslides and flooding management, and ecological conservation. This study aims to contribute to LiDAR's utility for urban analytics by overcoming current limitations of grid-based methods while enhancing classification in complex terrain. This research highlights the importance of LiDAR in making sustainable urban landscapes and beyond, significantly informed by data.