Multispectral Imaging for Agriculture FAQ
Frequently Asked Questions
What is multispectral imaging?
The colors we see in light are defined by the wavelength of that light. Plants absorb and reflect light differently depending on this wavelength. Plants typically absorb blue light and red light, while reflecting some green light. They also reflect a much larger amount of near-infrared (NIR) light, which is not visible to the human eye but is visible to the RedEdge™. The reflectance curve of a typical plant is shown below. Reflectance is the percent of light that is reflected by the plant.
By measuring the reflectance of a plant at different wavelengths, multispectral imaging enables identification of areas of stress in a crop, and provides a quantitative metric for the vigor of a plant.
How do multispectral cameras work?
Multispectral cameras work by imaging different wavelengths of light. The MicaSense RedEdge™ multispectral camera has 5 imagers, each with a special optical filter that allows only a precise set of light wavelengths to be captured by that imager. Once processed, the output of the camera data is a set of images where the value of each pixel is equal to percent reflectance of light for that particular wavelength. These sets of images are then stitched together to create geographically accurate mosaics, with multiple layers for each wavelength. Mathematically combining these layers yields vegetation indices. There are many types of vegetation indices that measure different characteristics of a plant. Some indices, for example, are useful for measuring chlorophyll content of plant leaves. Other indices can be used to estimate nitrogen content. Other indices provide indications of water stress. One popular index is the Normalized Differential Vegetation Index (NDVI), created by combining the reflectance from red and NIR light.
How does RedEdge™ differ from a single-imager multispectral camera?
A single-imager multispectral camera uses a blocking filter, combined with the standard camera's built-in filter to capture information in 3 wavelengths of light. Because these imagers aren't optimized for remote sensing, the built in filters are wideband and suffer from data contamination from neighboring bands. This figure shows a comparison of the RedEdge™ narrowband optical filters and the filters used in a typical single-imager camera.
The RedEdge™ camera uses narrowband filters with known characteristics combined with factory calibration parameters, enabling accurate measurements of reflectance that a converted camera simply cannot match.
The Micasense RedEdge can be purchased in Australia from Rise Above here