VL3D++ has been used to analyse point clouds in very different contexts: from full-territory classification to anatomical analysis in medicine.
In every case, the process is the same: a raw cloud, a Deep Learning model and a classification that can be measured, analysed and exploited.
All the examples shown below use exactly the same framework.
What changes is the type of point cloud and what you want to classify in it.
Automatic characterisation of terrain in riverbeds and floodable zones
From a point cloud of the fluvial environment, VL3D++ enables automatic classification of the terrain and its geomorphological variations along the watercourse, including riverbeds
Our technology automatically separates natural terrain from machinery and infrastructure. By accurately identifying slopes and structural break points, we generate high-fidelity digital elevation models for the design of mining benches and safe ramp planning. We detect and quantify earth movement through differential analysis of clouds captured at different points in time. This makes it possible to calculate excavation and stockpile volumes with absolute precision, optimising transport logistics and acting as an early warning system for geotechnical risks on slopes.
3D anatomical segmentation of dental and vascular structures
From 3D scans of anatomical structures (obtained with photogrammetry, intraoral scanner or volumetric reconstructions), VL3D++ automatically classifies the different parts of the anatomy, turning geometric data into detailed segmented information.
The largest territorial classification ever carried out with Deep Learning on a point cloud. We transform billions of points from the Spanish National Aerial Orthophotography Plan into a structured geographic database. Using artificial intelligence algorithms trained for the specific morphology of the northwestern peninsula, we accurately segment natural terrain from the various vegetation layers and artificial constructions. This automated categorisation makes it possible to filter noise to generate Digital Terrain Models (DTM) with a resolution unattainable by traditional methods, optimising territorial management, forestry planning and cadastral taxation.
Corridor control and detection of safety-distance violations, at massive scale.
Spanish regulations on high-voltage power lines (RD 223/2008 and ITC-LAT) set minimum safety distances, easement zones and clear limitations on trees and vegetation around power lines.
In addition, in some autonomous communities, fire-prevention legislation requires the management of plant biomass in the so-called biomass management strips near infrastructure, including power networks.
Forest inventories and yield estimation per tree and per plot.
From an aerial point cloud, VL3D++ automatically classifies vegetation and analyses its height, density and distribution over large areas.
Classification with VL3D++ enables individual tree counting, identifying the exact position, height and crown diameter of each specimen, turning the raw 3D point cloud into useful information for accurate estimates.
Collaboration, research and technology transfer
We foster an ecosystem of open innovation and continuous collaboration between academia and industry. Our technical and research team can guide you to explore how to apply VL3D++ to your case, collaborate on research or study technology-transfer pathways. Anyone interested in professional point cloud classification, or who needs the development of a specific technical functionality for their workflows, can contact our development team directly. We are open to joint research projects, technology transfer and advanced support for the implementation of the framework in production environments.
