It’s easy for people and most animals to tell the difference between ripe and unripe strawberries — just look for the red ones. But for robots, that’s no simple feat. So researchers at the National Physical Laboratory have developed a four-part technology to teach robots how to pick only the sweetest berries.
Ever wonder why strawberries start as whitish-green but suddenly turn bright red as soon as they’re ripe? It’s an evolutionary response to prevent birds from picking at the fruit before the seeds have fully matured. Green berries are much tougher to see against the green foliage than red ones are — it’s the biological equivalent to putting up a flashing neon sign.
The new robotic equivalent uses a mix of radio frequencies, microwaves, terahertz and the far-infrared to identify pickable berries through their foliage — literally peering inside the strawberries themselves and checking their ripeness based on predetermined indicators. For example, according to project leader Richard Dudley, “This is a fairly easy fruit to measure as it has high water content and dry leaves, and microwave imaging is particularly useful for identifying water levels.”
To determine what these criteria should be, the NPL set about exhaustively collecting measurements of the crop in both fields and laboratories to create a statistical measurement range. These measurements were then employed to build sorting algorithms to direct the scanning system towards only the ripest berries.
And since the method is completely non-invasive, it does not contribute to the thousands of pounds of waste generated from picking unusable strawberries. As Dudley points out, “strawberries are a high value fruit which are very time-consuming to pick, so there is a stronger business case to implement automated picking technology for strawberries.”
The system is also being adapted for use on other crops — it was originally developed in 2009 for measuring cauliflower ripeness — as well in other industrial uses such as sorting wood, paper and plastic items in recycling centres.
Monster Machines is all about the most exceptional machines in the world, from massive gadgets of destruction to tiny machines of precision, and everything in between.
[NPL, Gizmag, Computer Vision Central]
Top image: Igor Borodin/Shutterstock