Low-cost imaging platform for measuring wheat growth response to climate warming

Overview

Wheat is highly sensitive to temperature, and small changes in day/night temperatures can significantly alter growth rates and the timing of yield-determining developmental stages. Monitoring these temperature-driven checkpoints is essential for identifying breeding targets and managing crop inputs such as nutrient application. This project delivers a low-cost, field-deployable imaging platform that combines accessible hardware with machine learning to track wheat development under varying temperature conditions. The research also generates genetic insights through gene expression analysis, providing breeders with actionable targets for developing climate-resilient wheat varieties.

Technical specifications

Imaging platform:

• Raspberry Pi-based camera system capturing regular time-lapse images of plant growth
• Python-based image analysis pipeline for detecting and quantifying green plant pixels
• Designed for deployment in both controlled glasshouse environments and open field conditions

Temperature response analysis:

• Plants grown under varying day/night temperature regimes to assess growth rate variation
• qPCR analysis of leaf tissue to measure expression of candidate genes involved in temperature regulation
• Planned RNA-sequencing approach to capture total gene expression across a 24-hour day under different temperature and photoperiod conditions

Machine learning integration:

• Planned development of ML models to move beyond simple pixel counting and physical measurements
• Detection of individual leaf growth, total plant height, and other fine-grained developmental features
• Identification of critical developmental stages susceptible to climate warming

Genetic validation pathway:

• RNA-seq data used to identify key genes and gene networks governing temperature-regulated growth
• Mutational studies using existing SNP-based TILLING lines
• Potential for CRISPR knockouts or overexpression studies where budget permits

Technology readiness level

The imaging system has been established and validated in glasshouse settings, with initial results demonstrating significant growth rate variation under small temperature differences. Current growth measurement relies on physical measurements and basic green-pixel detection through Python-based analysis. qPCR has confirmed expression changes in candidate temperature-regulation genes. The next phase will introduce machine learning for more precise growth quantification and RNA-sequencing for comprehensive gene network discovery. The platform is positioned at an early-to-mid technology readiness level, with strong potential for field deployment and breeder adoption once ML-based analysis and full transcriptomic validation are completed.