Hyperspectral imaging and analysis represents the pinnacle of remote sensing and provides advanced solutions across a variety of scientific disciplines including forestry, vegetation management, infrastructure assessment, and water quality.
Hyperspectral Imaging Defined
Hyperspectral means “combining information from a large number of contiguous spectral bands of visible, infrared, and short-wave infrared frequencies,” according to its formal definition. It simply means we are capturing an image of the sun’s reflectance within hundreds of narrow bands of energy. This is unlike more traditional cameras, which measure the reflectance within three or four broad bands.
The applications of hyperspectral imaging for our clients are also varied and quite important, mainly in the broad areas of:
- Land Use Planning
- Aquatic Monitoring
This includes vegetation classification, vegetative health, invasive species, forestry, wetlands, water quality, bio habitats, agriculture, geology, exploration, and coastal science.
How Do Hyperspectral Imaging Sensors Differ From Traditional Cameras?
Our traditional digital cameras are classified as multispectral. With these we collect four distinct bands in the red, green, blue, and near-infrared regions. The spectral range is almost identical to the CASI 1500h capabilities, but instead of the four distinct channels found in a traditional multispectral camera, the CASI 1500h can capture up to 288 narrow channels within this same spectral range.
The true value of this technology is unlocked when we fuse hyperspectral data with information from other sources, such as high-resolution lidar point clouds. In vegetation management applications, we often use lidar to group elevation returns to model the structure of each individual tree in the AOI. We then attach the 2D hyperspectral classifications to the 3D tree representation to classify and assign tree species.