The first product produced by Bitwise Agronomy is Greenview, which combines computer vision, machine learning and artificial intelligence to count and measure components of horticultural crops, such as berry and bunch number, shoot length and/or fruit ripeness. These parameters underpin operational decisions by providing the data necessary to make data-driven decisions, and also to achieve the agricultural holy grail of yield estimation.

Market demand is strong to resolve the logistical challenge of inaccurate yield estimates, which includes not only processing constraints, but also labour hire, shipping and market supply. The accuracy of humans when performing simple, but repetitive tasks, such as counting bunches of fruit, can be as low as 50%. Greenview uses off-the-shelf cameras, such as GoPros, to capture the videos. We run our Bitwise Agronomy machine learning models on top of the video to identify, count and measure those parameters growers need to base decision making on.

The product was developed to first solve the problems of one of our co-founders, Fiona Turner. As the co-owner of Jinglers Creek Vineyard, Fiona needed data to make decisions such as spraying, lifting wires, fruit thinning and canopy trimming. Currently much of the imagery offered to gain insights into horticultural crops relies on aerial-based imagery. This presents a challenge with the resolution and the cost, and the Greenview product has resolved this by taking side-on images on a per plant basis. This gives the us the advantage of being able to actually ‘see’ and count the fruit, a basic yield estimation requirement, rather than model the numbers. Utilising commodity cameras has also enabled us to onboard customers globally from the beginning, even during a pandemic.