Canola is one of the world’s most vital oilseed crops, important for food, feed, and renewable energy. However, climate change threatens yields through increased drought and heat stress. Improving water-use efficiency is crucial, and stomata, the tiny pores on leaves that control gas exchange, are a key breeding focus. Up to now, measuring stomatal traits like density, size, and pore dynamics has been slow and labour-intensive, hindering progress in breeding for climate resilience.

To address this issue, researchers at ANU led by Dr Florence Danila developed SCAN. the Stomatal Comprehensive Automated Neural Network (SCAN). SCAN is machine learning-assisted image analysis suite designed for canola. SCAN can accurately estimate stomatal density, size, and pore area compared to traditional methods, and it can even detect the “open/closed” status of pores in real time. The researchers applied this tool across different leaf positions in the plant canopy and under various field and glasshouse conditions, demonstrating that stomatal traits vary with leaf age, environment, and different canola types. As an offline application, SCAN offers immediate results and high-throughput capacity for crop breeding programs.

Using SCAN, the research team showed how stomatal traits differ between leaf surfaces, canopy positions, and canola ecotypes under various environmental conditions. These insights provide valuable targets for breeding drought- and heat-tolerant varieties.

The impact of SCAN goes beyond a single project. It has opened the door for two major initiatives: a five-year GRDC project on source-to-sink relationships and genetic diversity in canola (2025), and an ARC Linkage Project focused on increasing yield through improved pod photosynthesis (2026). Together, these projects will utilise SCAN’s capabilities to speed up genetic improvements and achieve significant increases in yield potential.

By automating a process that used to take a lot of time, SCAN shows how advanced digital tools can improve crop development. It highlights the use of machine learning, field-ready technology, and plant science to build climate resilience and keep productivity high in one of Australia’s most important crops.

You can read more about Dr Danila’s research in SCAN: an automated phenotyping tool for real-time capture of leaf stomatal traits in canola | Journal of Experimental Botany | Oxford Academic.

Partner with us to future-proof your operations. Whether you're seeking strategic insights, technical solutions, or collaborative innovation, we’re here to help you lead in a climate-smart agricultural future.

Work with us