Camera Design Aims to Make Multispectral Imaging Cost-effective

Researchers from Friedrich-Alexander Universität Erlangen-Nürnberg (FAU) have incorporated three resolutions into a single camera using a cost-efficient process and construction. The intelligent device achieves high spatial, temporal, and spectral resolutions, making it suitable for enhancing human perception across applications and environments.

In the development of cameras, the researchers said, spectral resolution, which considers a captured image’s wavelength (and, hence, perception of colors), has been largely adjusted to match the capability of human sight. It has, therefore, traditionally corresponded to measuring red, green, and blue. The light spectrum extends beyond those colors, though, enabling additional tasks and functions that occur organically in nature or that can be created.

“Up to now, there were only extremely expensive and complex methods for measuring the ultraviolet or infrared ranges of light or individual spectral bands for special industrial applications,” said Juergen Seiler, part of the research team. “We looked for a cost-efficient model and we were able to develop a very cost-effective multispectral camera.”

The researchers, based at the Chair of Multimedia Communications and Signal Processing at FAU, connected multiple standard cameras with various spectral filters to construct a standard multispectral camera array. The researchers then calculated an image, combining the various spectral information each sensor delivered.

The concept allowed the researchers to determine the materials of an object captured with a single image. Further, as several “eyes” of the camera record the captured surroundings, as happens in human sight detection, the system also delivers an indication of image depth. Beyond material properties and color, the camera system additionally captures info on the distance between imaging system and object.

Potential applications for the new device and system include those aimed at environmental protection and resource conservation, as well as those in modern agriculture.

“A whole range of solutions to various problems has now opened up thanks to our new technology,” Seiler said. “In the infrared range, for example, we can differentiate between real people and signposts using the thermal signature. For night driving, we can detect animals crossing the road with sufficient warning.”

The detection and sorting of plastics, which can differ in/according to various ranges of the spectrum, is another possible application for the intelligent camera that the research team has identified.

The research was published in IEEE Transactions on Image Processing (www.doi.10.1109/TIP.2020.3024738).