Multispectral Images

A color-blind person cannot perceive some things because he lacks certain contrasting color receptors. Someone who is not color blind does not see everything either. In nature there are animals that can distinguish many more different wavelengths of light separately from each other such as for example the praying mantis shrimp which itself also has beautiful colors. This multispectral sight would not have evolved if there was nothing to see in these extra wavelength bands.  

Visible light (VIS)

Visible light ranges from 400 to 750 nanometers. We humans see color within that part of the spectrum by measuring the intensity ratios of red, green and blue (RGB). A color camera also records this in the same way. However, the RGB bands do not exactly match those of the human eye.

Near Infrared (NIR)

Just outside the visible light spectrum we find the near infrared spectrum. This is infrared light and cannot be seen with the naked eye but with the help of a camera it can. CCD and CMOS cameras have a spectral response that run from 400 to 1000 nm. When equipped with dedicated filters, it becomes a dedicated detector. Our camera system uses multiple CCD sensors with dedicated filters that capture images at different wavelength bands in sync. Such a setup is the basis of a multispectral camera setup.   

Short Wavelength Infrared (SWIR)

The near infrared spectrum is followed by the Short Wavelengh Infra Red (SWIR). Images in this band are usually detected with an InGaAs detector. A typical image property that can be observed in SWIR images is the high absorbtion of water. This strong absorbtion of water results in high contrast images that have water in it and can therefore be used to detect soil moisture. Just like in the VIS and NIR optical bandpass filters can be utilized for specifiec applications.  

Thermal Infrared (TIR)

A good bit further down the electromagnetic spectrum we find the Thermal Infra Red (TIR). With a thermal imaging camera we can capture it and translate it into an image. Thermal images can help detect live stock, fire hazard hotspots, but also can provide vital insights in irrigation. Plants that have plenty of water will stay cool on a sunny day, those that don't will get hotter as they don't have enough water to vaporize. 

Multispectral Map example 1, Tulip Fields:

Our multispectral maps contain multiple predefined wavelength bands. We combine general RGB color imaging with several near infrared bands. To do so we use a synchronized multi camera setup with spectral filters in front of the lenses. Depending on your application we can mount the desired wavelength bands.

Example 2, Soil moisture differences in peat meadows: