Towards real-time precision crop management: automatic during-harvest assessment of apple yield

点击翻译按钮获取中文摘要

After harvest, apples are stored prior to being graded, packed and shipped to retailers, but producers often have little reliable information on the stored crop, which makes it difficult to match customer requirements. This leads to wasted crop at a cost to the farmer, as well as wasted money from storing apples that do not match specifications in cooled stores. If we could produce a size profile of the crop at the time of harvest then this would substantially aid planning, save money and reduce food waste. Together, the project team will develop significant functionality into an existing crop sizing/counting system that is currently parameterised for potatoes. This will allow us to create an innovative machine-vision system to provide real-time, geo-located count, size, and marketable yield during apple harvesting. This system will involve no additional labour requirements, unlike other options (e.g. drones), and will increase productivity by ensuring apple specifications can be optimized to the crop, increasing our farms' and the sector's resilience. Less waste will reduce environmental impact and will help the drive to net zero emissions.