Discover how artificial intelligence is reshaping the future of agriculture, enabling scientists to accurately predict crop yields and overcome the challenges posed by climate change.
Empowering Plant Breeders
Plant breeding is critical to developing avenues for high-performing, more resilient crop varieties in the ever-changing agricultural landscape. This is far easier said than done, as accurately predicting crop yield is a difficult task.
Until, that is, the age of artificial intelligence (AI) and all its magic. Well, with the help of artificial intelligence and a recurrent neural network model, researchers at Purdue University have disrupted this dominance.
An AI model that combines data from diverse sources (such as remote sensing, weather and even genetics) for a much better understanding of plant breeding. By searching for spatial and temporal patterns across this data, the model can predict the yield of different maize hybrids with high precision at various locations and times so that breeders can make scientifically-endorsed decisions in creating resilient crops in preparation for climate change.
Plant Phenotyping Transformation
Phenotyping, the measurement of plant traits, was traditionally a costly and labor-intensive process. These time-consuming logs, from getting plant heights to chemical analyses, have been a bottleneck for agricultural research.
But the use of remote sensing technologies like drones (uncrewed aerial vehicles or UAVs) and satellites has made it easier. Hyperspectral cameras take much more detailed reflectance measurements of light wavelengths beyond the visible spectrum, while Light Detection and Ranging (LiDAR) measures plant structures in 3D.
Helped along by these advanced remote sensing technologies and deep-learning algorithms, scientists have created a new, readily accessible tool for identifying healthy crops compared to stressed ones faster and more accurately than human eyes. This not only saves on labor costs, but it also alerts growers to important data that can be used to make their management practices are as efficient as possible in order to improve crop yields and resilience.
Conclusion
The use of artificial intelligence and remote sensing in combination with agriculture is a paradigm shift to describe future perspective in the capture and prediction of crop yield. They are employing deep-learning models with a host of additional data — genetic, environmental and plant phenotypic to support better-informed decisions by plant breeders and growers, boosting food security and sustainability in the age of global climate change. This stands to reason as the technology matures further we will see leaps and bounds in this realm of precision agriculture, ultimately moving towards a future wherein the success of our entire global food systems hinges on AI-derived intelligences.
