The aim of Sense.Lidar® is to improve the lidar classification accuracy and quality so that more automated analysis can be performed when quality data exists (both historical and newly collected).
Key Features:
- Accurately classifies lidar point clouds or clusters of points to best represent the feature (for example: buildings, trees, assets) to 95%-99% accuracy as desired.
- The accurately classified lidar data further improves lidar product development like building footprint polygons, 3D building models, vegetation density and type, etc.
- Efficiently and accurately assists with creating lidar-derived 3D digital twins of the human- and natural-built environments
- Improves the creation of digital elevation models (DEM) by generated hydro- and building-flattened DEMS for accurate flood analysis
- Can be localised, or scaled to city, county, or country-wide analyses through cloud-based processing
- Machine learning enables the algorithms to get better over time
- Results can be delivered via a 3D (web) environment or integrated into client GIS systems
3D buildings
Sense.Lidar® accurately classifies buildings to support the development of building footprints and 3D building models. Through machine learning techniques, the processes are trained to accurately identify the building and separate it from neighbouring features (like trees, cars, and bushes). The output is a clean and accurate cluster of classified points that are used for creating a building model that maintains important details like roof height, pitch, area size, etc.
Vegetation
Sense.Lidar® accurately classifies vegetation to support the development of vegetation height and density models for better measuring the geolocation, characteristics, species and volume of shrub, bushes, and trees.
Improved digital elevation models (DEM)
DEM are enhanced with the added feature of building flattening. It is most common for a DEM to be hydro-flattened or enforced to support water flow analysis. Adding building flattening utilising a building polygon created from an accurately classified lidar point cloud further enhancing the DEM by flattening the surface that exists when removing a building from the DEM. This helps to improve water volume analysis by removing areas of artificial pooling caused by removing a building and not flattening the surface where the building exists.
Asset management
By accurately classifying features using Sense.Lidar® asset managers can now geo-locate recognisable features from the varying clusters of points. The lidar cluster density would determine the recognisable feature. For example, a low-density product (2 points per square meter) is adequate for identifying buildings, trees, bridges. A high-density product (30+ points per square meter) can identify power lines, poles and utility features.
Land characterisation
Since Sense.Lidar® is scalable, classifying large areas accurately provide opportunity to get the big picture of land use, type, or category. These clusters of classified points can assist in making maps that characterise the earth’s land cover or land use, identifying urban, suburban, or rural areas, or various coastal zones and wetlands.