Hyperspectral imaging reveals enemy targets
Hyperspectral technology is based on the differentiation and identification of objects based on how they reflect the different wavelengths of light. The image recorded by an ordinary digital camera consists of information from the three wide bands of visible light. A hyperspectral camera, on the other hand, records information on dozens or even hundreds of narrow wavelengths.
The spectral image generated by the method is a three-dimensional cube of data, consisting of several superimposed black-and-white or colour images. When a spectrum is determined for each pixel, the individual spectra of different substances can be used to identify different targets. The analysis of the data requires algorithms.
Oulu-based Senop develops and manufactures hyperspectral cameras for the needs of industry, science and defence. The business is based on the long-term product development of Rikola Ltd Oy, which was acquired by Senop in June 2016.
In its hyperspectral cameras, the company utilises Fabry-Pérot interferometer technology, principally studied in Finland by VTT Technical Research Centre of Finland. The technology involves adjustable optical filters that enable the cost-effective manufacture of small but powerful devices.
For example, the hyperspectral camera developed for drones only weighs 720 grammes.
“In the international market, we are particularly well known for our cameras suitable for unmanned aircraft”, says Jussi Soukkamäki, Sales Manager in charge of RIKOLA products at Senop.
The history of hyperspectral cameras dates back to the 1970s. The technology was first put to military use, but has found several civilian applications in recent years.
For example, hyperspectral imaging has been used for remote mapping. A camera mounted on an aircraft or satellite can chart forest cover or determine the amount of nutrients in a field, enabling the precise application of fertiliser.
“In the future, hyperspectral imaging will see more widespread use in the detection of plant diseases and weeds. It can also be used to analyse the water economy and irrigation needs of fields”, Soukkamäki says.
The utilisation of hyperspectral technology has also extended into the field of industrial process and quality management. In addition to these, there are major applications for the technology in medicine.
“For example, a hyperspectral camera can identify premalignant conditions on the skin”, Soukkamäki points out.
Senop’s unit manager in Oulu Jani Mäklin thinks that hyperspectral technology will also see increasing use in military applications in the future. It can be used to detect mines or roadside IEDs, for example.
“Digging a mine into the ground will always leave traces, even if they are undetectable to the naked eye. The solution to this problem is a hyperspectral camera with a good resolution”, Mäklin says.
Hyperspectral imaging can also be used to separate camouflage nets from vegetation. In addition, the technology is capable of detecting whether an armoured vehicle concealed in terrain is real or an inflatable decoy.
“A hyperspectral camera can also be used to study the shape and humidity of terrain. This can be used to determine whether a given patch of ground is able to support a vehicle”, Soukkamäki says.
The compact technology is also suitable for hand-held devices. According to Mäklin, future applications include high-resolution binoculars whose image can be tailored to the targets you are looking for.
With the press of a button, such binoculars will show camouflage webbing, while pressing a different button will show armoured vehicles in the same image.
“RIKOLA cameras enable the quick and precise selection of spectral bandwidths according to the desired target. This lets users limit the amount of data displayed, which makes it quicker to analyse”, Mäklin points out.