A new research project at Lethbridge College is using near-infrared (NIR) hyperspectral imaging technology to test potato quality without having to cut tubers open, a June 16 news release said.
Traditionally the best way to test a potato’s quality is to cut it open and see if there are any defects. This process is time-consuming and destructive, taking good potatoes off the production line for random testing, the release said.
The new three-year $523,300 research project is being led by Chandra Singh, senior research chair in agricultural engineering and technology at Lethbridge College. It’s being funded by more than $400,000 from Results Driven Agriculture Research and Alberta Innovates.
The project is using NIR hyperspectral imaging system and machine learning techniques to detect quality parameters in potatoes, without destroying the samples. The release noted most major quality issues associated with potatoes are internal defects, greening, specific gravity and sugar content.
“The cost has gone down for this technology, the cameras and supporting technology are less expensive and the processing is much, much faster. So, we are at a point where this technology can be implemented on a large-scale,” Singh said in the release.
The release said the college had previously acquired the technology that will be used in the project through a Natural Sciences and Engineering Research Council of Canada grant in 2021.
The project will identify the most significant wavelengths required to detect the quality parameters associated with potatoes, the release said. It’ll also test the NIR hyperspectral imaging system at speeds simulating commercial scanning speed and will eventually design a prototype for commercial application of the system based on extensive testing and analysis.
Singh is collaborating with University of Lethbridge Research Scientist Michele Konschuh on the project, with funding contributions also coming from Lamb Weston, Old Dutch Foods and the Edmonton Potato Growers, the release said.