Czech researchers develop AI system to automate plant emergence monitoring

Researchers from CATRIN at Palacký University, in collaboration with the Czech University of Life Sciences Prague, have developed an AI-based system called SPROUT to automate the monitoring of plant seedling emergence.

SPROUT (AI-based Seedling PRedictiOn and trait extraction Using RGB Time-series) analyzes RGB images over time to detect exactly when crops emerge and extract biologically significant parameters. This replaces manual monitoring, which often only records final emergence percentages and misses critical data regarding the progression of growth.

The system was tested on barley and wheat data using Temporal Convolutional Network (TCN) models, achieving up to 90 percent accuracy in determining emergence timing. Designed as a modular system, SPROUT can be retrained for different crops, cameras, or experimental conditions, making it a potential tool for breeders, seed companies, and agricultural manufacturers.