A new AI system is being built and developed to improve how farmers predict and optimise their crop yields. With counting flowers on fruit trees using images taken from a standard smartphone, this technology brings farmers a smart, efficient alternative to traditional methods.
In Spain, where the system was piloted on peach orchards, it reached an accuracy of 90% in flower counting, a major improvement over the usual error rates of 30-50% seen with manual counting.
Why Does Accurate Flower Counting Matter?
Accurate flower counting leads to precise yield predictions, crucial for efficient farm management. These better and more accurate forecasts mean farmers can plan water usage, labour needs, and harvest logistics far in advance, ultimately minimising waste and taking up profitability.
Fernando Auat Cheein, an associate professor at the National Robotarium, discussed the practical benefits: “By integrating this technology, farmers can align their resources more closely with expected yields, which helps in reducing excess expenditure and environmental impact.”
How Was The AI Developed?
The AI system was developed through collaboration between tech developers and agricultural experts at the National Robotarium in Edinburgh. They made use of machine learning algorithms that could analyse thousands of images to learn and identify the intricate patterns of flower clusters, even when obscured or overlapped.
This way, its made sure that the tool was developed to adapt to real-world farming needs, for better usability and effectiveness.
More from Artificial Intelligence
- Experts Share: How Is AI Influencing Advertising?
- What Are Japan’s New AI Laws, And Does The UK Have Similar Policies?
- Reports Find That AI Skills Are Becoming Important In The Job Market
- What Is Google’s Vision AI, And What Is It Used For?
- What Cybersecurity Measures Are Available For AI?
- OpenAI Explores Developer Interest in Integrating ChatGPT Directly Into Apps
- How Reliable Are Artificial Intelligence Research Tools?
- The Rise of AI Recruiters & The Startups Leading The Charge
What Issues Did Developers Overcome?
Developing an AI that accurately processes natural elements like flowers involved overcoming a few things:
Complex Pattern Recognition: The AI had to distinguish between overlapping flowers and leaves, a task requiring sophisticated image recognition capabilities.
Variable Environmental Conditions: Changes in light, weather, and bloom stages required the AI to be highly adaptable and responsive to ensure consistent accuracy.
Integration with Farmer Routines: The tool had to be simple enough to integrate into farmers’ daily routines without disrupting their established practices.
Real-World Impact- What Are Farmers Saying?
Feedback from farmers who tested the AI has been for the most part positive. A peach grower from Spain, for example, shared, “It’s like having a futuristic tool in the palm of your hand that simplifies one of the most tedious parts of our work.” Many expressed excitement about the potential time savings and the reduced risk of human error in yield estimation.
Where Else Is AI Being Used In Harvesting?
Back in the UK, researchers at the National Robotarium in Edinburgh are applying AI to improve the management of resources like water, essential in farming. Accurate forecasts of harvest sizes allow for more precise water distribution, so that plants receive just what they need without wastage.
The technology is adapted to refine farming methods themselves. Farmers, in accurately predicting areas of higher yields, can concentrate their efforts on those sections for pruning and herbicide applications.
Fernando Auat Cheein, Associate Professor at the National Robotarium, explained, “This targeted approach helps us apply resources effectively, optimising the health and output of crops.”
These advancements show the flexibility of AI in improving traditional farming methods, assisting in the precise management of resources and refining agricultural practices.
