Skip to the content

Hyperspectral vertical images

Multi-colour images are called multi-spectral imaging, such as an orthophoto with red, green, blue and infrared channels corresponding to four ribbons. When moving from a few spectral channels to a continuous spectrum of several hundred channels, it is called hyperspectral imaging.

Our hyperspectral orthophoto covers the visible light spectral regions (400-700 nm), near infrared (700-1000 nm) and short-wave infrared (1000-2500 nm) in nearly 500 channels. With such accuracy one can uncover even small differences in the spectral signature in the surface, which can then be used for different classification.

The possible applications are almost endless. Urban planners, environmental researchers, forest owners, road managers and most people can access unique decision-making and planning tools.

Some examples of practical applications are:
* Automatic extraction of objects from a large dataset; e.g. asbestos roofs, solar panels, yellow cars, trampolines, rare plants, oil spills
* Overall classification of surface types: building, vegetation, glass roof, pavement, water
* Classification inside the subclasses: different materials in rooftops, different types of vegetation, different types of pavement
* Condition assessment of different subclasses: vegetation health, damage and wear on buildings, new / old asphalt, holes / cracks
* Water, seabed and beach zone: depths in water, classification of underwater vegetation, turbidity in water, algae concentration, heavy metals, oil spills
* Environmental monitoring: surface pollution and pollution in the air over the city

We can supply the datasets as a whole or offer separate analyses and classification as desired.