Fruit is a natural product that presents unique problems for computers
A coconut mik producing plant in Thailand paid farmers based on simple unit count as trucks were unloaded.
Counts were not accurate but more importantly, the size and grade of the coconut was not taken into consideration.
The objective was to accurately pay farmers for product delivered based on quantity, size and quality.
During the coconut season, trucks would arrive at the plant "24/7". Coconuts were unloaded onto a continuous conveyor.
The system had to be deployed on this conveyor.
The system had to keep an accurate count of coconuts by size increment. In addition it was necessary to identify coconuts that were damaged or had mold defects.
Once the aggregate result was obtained, the farmers account needed to be credited.
The iamge collection system was fairly simple but it was decided to use deep learning software to "grade" coconuts by size and defect category
The system was deployed into the production line. No operator intervention was required during normal operation.
The system did not require routine servicing other than daily cleaning of the camera lens.
The client achieved significant cost reductions in not having to deal with defective product. In addition, yields increased as farmers only brought high quality fruit to that facility.
To find out how Indyn can help with your particular application, just click the link belowdiscuss