Cutting-edge Analysis of fruit crops
IFV’s Fruitvision system will provide growers with the most comprehensive view of their crop possible. We have developed a suite of industry-leading solutions that capture and analyse millions of points of data and turning it into information that growers can retrieve in a clear and concise format, so that they can make more effective decisions.
A key part of management practice is to know when & how to ‘thin’ crops to promote selective growth. This is very dependent on knowing accurately how many apples there are on each tree in the growing season. We believe that with the ability to identify trends & potential defects earlier in the growth cycle, we can improve orchard management techniques. This will enable them to develop a precision agriculture approach towards increasing tonnage, Class 1 yields & thus profits, by improving vigour and nutrition balance through specific pruning tactics, fruitlet size-thinning strategy and precision fertigation.
To achieve this we need detailed information on growth & development & our system will enable growers to:
- Measure apple yield and size development in season, with accurate prediction of Class I yield
- Negotiate contracted volumes accurately without wasting Class I apples, or having to import to make up shortfall.
- Reduce ‘rejects’ by 50% upgrading more crop to Class 1 through improved precision in size thinning & better crop management.
- uplift yield from current ‘average’ (29T/ha) to near ‘best’ (37T/ha) through strategic management (pruning; thinning; fertigation)
- Achieve sustainable intensification – by increasing yield by 8T/Ha from the same number of trees
The detailed analysis that the algorithms perform requires high quality images. In an 8 hour day, our 4 camera system will gather 5Tb of data. Over a single 4G connection this would take 31 days to stream to a server (that’s 93 x longer than the recording time). Even streaming via Wifi over dedicated superfast broadband would take several days. Dedicated superfast Broadband Wifi connection per camera would be needed for real time streaming. This together with data storage constraints is why we choose to perform real time ‘onboard’ processing for our system which offers rapid & concise reporting of crop variables.