Apple Counting in Orchards
Task objectives can be queried UFMFRR-15-M Machine Vision Assignment_v2 document
The goal is to use two different deep learning methods to count apples in photos of apple trees in orchards based on MinneApple, a benchmark dataset for apple detection and segmentation, (Häni, et al., 2019) , for yield mapping of apples in orchards.
The task completion process can be found in document MV AS1 apple_counting
Dataset: MinneApple
Two vision algorithms: Faster-RCNN(Two-stage) and YOLOv3(One-stage)