The company is now incorporating AI into other aspects of the business, including automating their sensor and integrating generative AI into its computer-vision algorithms. This ensures the model can measure plant characteristics regardless of its surroundings, increasing its versatility and effectiveness. The team at Gardin is also publishing a paper on their work.
Key takeaways:
- Gardin, an agricultural technology company, uses a sensor to monitor plant health and generate growth insights but struggled to source images of diseased plants to train a disease-detection algorithm.
- To overcome this, the company used generative AI to create synthetic images of diseased plants, bridging the data gap and enabling the creation of a model.
- The generative AI model was developed in-house and took four months to develop, with success measured based on its classification accuracy.
- Looking forward, Gardin plans to incorporate AI into other aspects of the business, including automating their sensor and integrating generative AI into its computer-vision algorithms to ensure it doesn't overspecify on one type of plant.