SDSU Develop AI-Powered Tool to Improve Wheat Farming

By Scout Nelson

Researchers at South Dakota State University (SDSU) are developing WheatAI, an artificial intelligence-powered application designed to improve wheat breeding and help farmers make better crop management decisions. The platform uses smartphone and drone images to analyze plant characteristics, estimate yields, and identify potential diseases.

Professor Sunish Sehgal began collecting wheat images in 2022 at SDSU research farms near Pierre and Brookings, along with producer fields in Winner and Hayes, South Dakota. In 2025, he partnered with Maitiniyazi Maimaitijiang, an assistant professor in the Department of Geography and Geospatial Sciences, to develop a working prototype.

The project combines Sehgal's expertise in wheat breeding and genetics with Maimaitijiang's knowledge of remote sensing, artificial intelligence, and computer vision. With support from graduate students, the team has developed a platform that can analyze nine plant characteristics, with plans to add another 10 to 12 traits.

WheatAI is expected to become publicly available for free through wheatai.net, with a tentative launch planned for 2028. The platform aims to make advanced wheat analysis accessible to farmers, researchers, breeders, and students.

Improving Wheat Breeding Through AI

Traditional wheat phenotyping involves manually measuring plant characteristics, such as kernel numbers, tillers, and disease severity. This process can be time-consuming and produce inconsistent results because assessments may vary between researchers.

WheatAI uses automated image analysis to reduce human bias and improve the speed and consistency of plant evaluations. It can also help researchers examine stomatal characteristics and analyze larger numbers of wheat samples to identify promising varieties.

Supporting Farmers With Better Crop Decisions

The application could help farmers monitor field conditions and estimate crop performance more efficiently. Instead of manually counting wheat heads in selected areas, producers could upload 10 to 20 field images to receive automated estimates of wheat heads per square foot.

These assessments could help farmers make earlier decisions about whether to harvest crops for grain production or use them for silage. The platform may also assist in evaluating Fusarium-damaged kernels, which can affect wheat quality.

Future Development of WheatAI

The SDSU research team continues to expand WheatAI's capabilities ahead of its planned 2028 release. By combining artificial intelligence with smartphone and drone imaging, the project aims to improve wheat research, support crop monitoring, and provide farmers with more reliable information for agricultural decisions.

Photo Credit: south-dakota-state-university