Advancing turfgrass maintenance with synthetic data for divot detection

A divot is a small piece of grass or turf unintentionally dug out of the ground during sporting events. Proper care of turfgrass divots is crucial for ensuring the maintenance of sports fields and golf courses. Precision Turfgrass Management (PTM) employs a range of advanced technologies to provide sustainable and efficient care for turfgrass. Although deep learning offers a powerful method for data examination, it requires an extensive, labelled dataset. This research presents the creation of a photo-realistic dataset, crafted using Blender, tailored for turfgrass divot object detection. This synthetic dataset aids in training deep learning models to detect irregularities such as divots in turfgrass terrains.