Seeding success with machine learning and computer vision
Last updated: 10 Jul, 2020
Researchers have taught a new tool how to carry out one of crop breeding’s more challenging tasks.
Growers need seeds that germinate effectively and uniformly within a given time period to maximise crop productivity.
Seed suppliers must therefore test seed samples to ensure a certain germination rate is met, a process that is difficult and time consuming.
SeedGerm offers an easy-to-use, low cost and scalable solution to this problem using machine learning.
The product is the result of a collaboration between the Earlham Institute, the John Innes Centre, Syngenta and NIAB.
One of the co-first authors, Carmel O’Neill from the John Innes Centre, said ‘Currently most seed germination is still recorded manually. Against this, SeedGerm presents fast, accurate, high throughput screening and will be of major interest to crop seed production companies and research programs screening large germplasm collections’.
SeedGerm uses a cabinet equipped with cameras that take photographs throughout the germination process, documenting each stage from imbibition (seeds taking up water) through to the emergence of the root and further changes in the newly growing plant.
Supervised machine learning is used to automatically determine how germination is progressing through comparing images. For example, algorithms can be trained to predict how likely it is that a seed has germinated based on measurements extracted from an image that relate to the seed’s size, shape, and colour.
Figure 1 from the study showing two types of SeedGerm hardware. The full figure description can be found here.
Josh Colmer, a PhD student at the Earlham Institute and co-first author of the study, said ‘To apply machine learning so effectively to test seed germination marks an exciting step forward, especially as the learnings from this project can inform a variety of image based analyses with wide-ranging applications in crop research’.
Seed germination experts from Syngenta confirmed the effectiveness of SeedGerm for measuring germination rate and seedling health across five major crop species, including tomato and rapeseed. There is potential for SeedGerm to replace manual seed scoring, while also contributing to seed certification, seed insurance and sowing guidance.
The power of SeedGerm to measure phenotypic changes over time has further novel applications in crop improvement research. Many of the characteristics that can be measured help to estimate performance in the field in terms of canopy closure, weed suppression and predicted yield.
SeedGerm has already been used in a Genome Wide Association Study of rapeseed at the John Innes Centre. The study discovered a chromosome region that explained the difference between low- and high-germinating seeds. A gene in that region is related to one involved in abscisic acid (ABA) signalling in Arabidopsis, which offers an interesting target for future studies and highlights the usefulness of SeedGerm in measuring the effect of genetic changes.
Ji Zhou, formerly of the Earlham Institute and now Head of Data Science at NIAB, said ‘We are excited to work jointly with the Earlham Institute and the John Innes Centre to accomplish this seed germination research. It will provide next-generation solutions for both academic and industrial applications, including our NIAB LabTest services, so that we can supply impartial science-based advice and cost-effective help to seed testing and address related agricultural and horticultural challenges.’
‘We are keen to continuously develop our Agri-Data and scientific computing capabilities together with both leading institutes so that we can offer innovative solutions for the Agri-Food and the broader crop research community - strengthening the UK’s core competence in data science, Agri-Tech innovations, and AI-based crop informatics’.
Rene Benjamins, Senior Scientist at Syngenta Seeds said ‘The developments and learning from SeedGerm are truly a big step forward in automation and generating high quality and reliable data in scoring seed germination. This will help seed companies like Syngenta in providing the best quality to their customers’.
Read the paper: Colmer, J., O’Neill, C.M., Wells, R., Bostrom, A., Reynolds, D., Websdale, D., Shiralagi, G., Lu, W., Lou, Q., Le Cornu, T., Ball, J., Renema, J., Flores Andaluz, G., Benjamins, R., Penfield, S. and Zhou, J. (2020) SeedGerm: a cost‐effective phenotyping platform for automated seed imaging and machine‐learning based phenotypic analysis of crop seed germination. New Phytologist. doi: 10.1111/nph.16736