Multi-granularity calculation method for hybrid scene airborne laser point cloud classification

Abstract

The invention provides a multi-granularity calculation method for hybrid scene airborne laser point cloud classification. The multi-granularity calculation method comprises the following steps: selecting a classification neighborhood point set and a scene neighborhood point set by taking a sampling point as a center; training a point cloud global feature extraction model in an unsupervised learning mode to realize coarse-grained scene perception; a feature fusion strategy based on an attention mechanism is adopted, space context information is embedded in a point cloud semantic segmentation model, a multi-task loss function considering the terrain clearance and the category is defined, the category and the terrain clearance of each point in a classified neighborhood point set are supervised, and fine-grained point cloud semantic segmentation and terrain clearance prediction are achieved; through point cloud segmentation based on graph cut optimization and iterative adsorption of a ground triangulated irregular network, fine-grained ground classification result refinement is realized. According to the method, the classification problem of the mixed scene point clouds is decomposed into a combination of three relatively single problems, the complexity of the whole problem is effectively reduced, and robust and fine classification of the point clouds of different complex scenes can be realized.

Description

technical field [0001] The invention belongs to the field of laser scanning data processing, and in particular relates to a multi-granularity calculation method for airborne laser point cloud classification in mixed scenes. Background technique [0002] Airborne Laser Scanning (ALS) is an important means of rapid acquisition of large-scale 3D geospatial data. It plays an important role in major national needs such as smart cities, global mapping, and global changes. It has been widely used in cities In national production or scientific research such as 3D reconstruction, forestry survey, shallow sea sounding, relic archaeology and deep space exploration. As the key foundation of many applications of ALS system, point cloud classification has a wide range of practical value. However, until now, this technology still plagues the industry and academia (the problem has not been completely solved), and has become a key bottleneck restricting the efficiency and automation of 3D g...

AI Technical Summary

Problems solved by technology

[0006] Aiming at the deficiencies of the prior art, the present invention proposes a multi-granularity calculation method for airborne laser point cloud classification in mixed scenes, and decomposes the classification problem of point clouds in mixed scenes into "scene knowledge learning + knowledge-guided classification + ground misclassification correction" "The combination of three relatively single problems achieves robust and high-precision classification capabilities for point clouds in different complex scenes

Images