Mediaboss Marketing

CUYAMA, CA - APRIL 28: Overhead irrigation of this newly planted crop of carrots is putting ... [+] pressure on the available groundwater supplies as viewed on April 28, 2020, in Cuyama, California. Located in the northeastern corner of Santa Barbara County, the sparsely populated and extremely arid Cuyama Valley has become an important agricultural region, producing such diverse crops as carrots, pistachios, lettuce, and wine grapes. (Photo by George Rose/Getty Images)

The globalprecision farming marketincludes technology like robotics, imagery, sensors, artificial intelligence (AI), big data and bio-engineering is expected to reach more than $16 billion by 2028, according to aMarch 2021 reportfrom Grand View Research.

What if you could combine AI and traditional aerial imagery to build data sets that help farmers and food processors gain insight into crop heartiness while it was still growing in the field?

Saul Alarcon, an Agronomist atThe Morningstar Companythat sources and processes tomatoes for several tomato-based products, says that new agriculture technologies based on AI can improve farming decisions. "Accuracy and consistency of data are very important to minimize the impact of crop's yield-limiting factors," said Alarcon.

"Smart farming technologies are becoming, in a short period of time, a key alternative in our worldwide efforts to improve the quantity, quality and nutritional value of food," said Alarcon. "Similarly, we firmly believe that it offers great opportunities to improve our environment while helping farmers to remain profitable."

John Bourne, vice president at Ceres Imaging, says that because food processors are increasingly using AI-powered aerial imagery to help manage their operations, they can now apply that to yield forecasting, quality control and risk mitigation.

"Typically processors pay for imagery and then offer the imagery service as a benefit to growers in their networks at no cost or for subsidized pricing," said Bourne. "This benefits the growers because they get reduced price imagery and product quality control vetted by their processors."

Images paint a picture, but AI images can help provide actionable data for farmers.

"Convolutional neural networks are used to enhance the accuracy of indexes such as segmenting images to identify pixels that belong to the crops we're measuring, and excluding all soil, grass and shadow," said Bourne. "AI can also classify individual plants and the pixels that belong to those plants."

But Bourne says that convolutional neural networks are also used to go from an index to a recommendation for a farmer, which means they could better identify certain acute irrigation issues, such as malfunctioning sprinklers with pins dropped in the imagery and ranking in terms of severity and risk to yield.

Patrick Tokar, Viticulturist atRombauer Vineyardsin Napa Valley, says that the vineyard initially looked into aerial imagery because they were searching for another tool to determine their irrigation needs. The company used Normalized Difference Vegetation Index (NDVI) to help determine the density of a green area in a patch of land but ended up at Ceres Imaging to address irrigation.

"This technology enables us to view the relative water stress for an entire vineyard block as opposed to specific data points within a block," said Tokar.

"What we did not realize when we first started using the service is the amount of correlation between water stress areas and wine quality," said Tokar. "We have traditionally used only NDVI images to map out harvest zones, but given our experiences over the past few years, we now look at the water stress maps in conjunction with the NDVI's."

The aerial data that Ceres processes is transformed into indexes that tell a different crop or yield story based on that index, such as water stress.

"Instead of looking at specific data points in the field to make decisions, aerial imagery gives us literally a bird's eye view of the entire vineyard block," said Tokar. "This enables us to hone in on any problem areas we may not have been aware of otherwise.

Tokar says that by looking at the imagery data, they saved time by planning out specific areas they needed to look at before a site visit, rather than scouting an entire vineyard to find potential problems.

Bourne adds that the primary driver for achieving a high solid percentage optimizes the farmer's irrigation strategy.

"Our most popular index is our water stress index which measures crop transpiration or how much a crop sweats," said Bourne. "The farmer can use the information from the index in several ways such as identifying irrigation issues like clogs and leaks in irrigation equipment."

Bourne says that when they publish the water stress index, the data is passed through an algorithm using convolutional neural networks to look for stress patterns. "The system can then identify issues and predict with a high degree of confidence the cause and severity of such issue such as identifying a grower has an irrigation pressure issue, that impacts six acres with high severity impact on yield," added Bourne.

Bourne adds that farmers can make adjustments caused by human error - an irrigation valve left on, equipment malfunctions, blocked irrigation nozzle, and even optimizing the irrigation schedule.

"For example, aerial imagery could show that the farmer has underwatered or overwatered a parcel of land, or it could show that one section of a block needs more water and one needs less water. So from this, the imager can make zone maps to facilitate watering that fits these issues," said Bourne.

Alarcon says that aerial imagery provided them with high-resolution images of the row and permanent crops. "This technology gives us the advantage of a wider spatial detection of potential yield-limiting factors in crops," says Alarcon.

"Yield uniformity can be improved by assessing low vigor areas during critical crop production stages. Factors like non-sufficient water levels due to low water pressure, plugged-up emitters, insects and disease damage, etc., can be rapidly detected and corrected through the use of crop aerial images.

Bourne believes that this knowledge lets the farmer "dial in" what they want as a result.

"By example, in tomatoes, a common metric is solid as a percentage of total tomato weight its water as a percentage of total tomato," said Bourne. "The grower gets paid more for high-quality tomatoes, so a high solid content tomato generally tastes better can be used fresh for things like salsa which is a higher value and a higher margin use."

Ceres Imaging is based in Oakland, California, and hasraised $35Mto date from institutional investors, including Insight Partners and Romulus Capital.

Originally posted here:
This Winery And Tomato Processor Used Artificial Intelligence To Make Their Crops Better - Forbes