A company is developing a computer vision system to assess vineyard crop yield with the joint effort of UC Davis scientists.
On Friday, Nathan Strong, president, and chief executive officer at Terroir AI in Menlo Park, exhibited the system at UC Davis’ Oakville Station. Strong introduced a case containing four sensor cameras in the back of the station’s ATV, the cameras adjusted for the high-wire trellis system of the experimental block.
The sensors collect data on the number of clusters and volume per push while the ATV or tractor travels up to 5 miles-per-hour; machine learning algorithms at that point gauge harvest yields. “The computer vision discloses to you what you have at the present time,” Strong said.
Terroir AI in the second period of improvement needs to manufacture man-made reasoning models to predict yields a long time in front of the harvest, considering farming management rehearses, the environment, and different components, Strong said. This will enable ranchers to meet their objectives. In the end, the units and algorithms will be intended to spot mites for disease detection.
Kaan Kurtural, the associate cooperative extension expert in viticulture at UC Davis, has planned the 1-acre completely motorized vineyard in Oakville, where Strong has tried the Terroir AI unit. The block incorporates 1,349 vines that are separated 1.5 meters by 2 meters – or around 5 feet by 6.5 feet.
Kurtural worked together with Strong and his group to improve Terroir AI’s computer vision system, including making sense of an approach to deal with occlusion from leaves, a typical issue found in other mechanized crop yield systems. Terroir AI’s technology resolved the occlusion issue to some degree by positioning cameras inside the enclosed unit to shoot the vines from various directions.
