Automatic Harvest Quality Monitoring

This project aims to measure the quality of corn silage harvested with a forage harvester. A main indicator of quality is the amount of fragmentation of corn kernels. The kernels should be sufficiently cracked such that the starch content is easily accessed by dairy cows when the silage is used as fodder. Current methods require that the farmer separate stover (leaves & stalks) from the kernels before a quality measurement can be made, however, this can be cumbersome and prone to error. Therefore, this project aims to directly measurement the quality of non-separated kernel/stover samples using a sensor such as an RGB camera. Focus areas in this project include: deep learning, object detection and segmentation.