Dynamics Modelling of Crop Growth from Images & Video Sequences in Protected Cropping

About the Project

This project will appoint one HDR student as part of the ARC Training Centre for Smart & Sustainable Horticulture collaboration between industry and research institutions to develop new cropping opportunities for the protected cropping industry. The project will contribute to the education and training of students and personnel for the protected cropping industry.

We are seeking a highly motivated PhD candidate to join an exciting research initiative within protected cropping systems (e.g. greenhouses, polytunnels, and indoor vertical farms). This project focuses on developing innovative, end-to-end machine learning models that learn and predict crop growth dynamics directly from sequences of raw images (possibly hyperspectral imagery) or video sequences, conditioned on daily environmental control inputs such as temperature, irradiance, water supply, and nutrient delivery.

Unlike traditional approaches that rely on extracted phenotypic traits (e.g., height, leaf area, or biomass proxies), this work treats visual appearance at each time step as the observable system state. By modelling the temporal evolution of plant images/videos under varying control inputs, the project enables powerful scenario forecasting, for example, predicting how crop development and performance would change if water supply were halved. The models will support a wide range of greenhouse crops and contribute to optimised resource use, stress resilience, and sustainable horticulture.

The project is fully independent, self-contained, and runs in parallel with complimentary imaging research. It offers strong potential for high-impact publications in computer vision, machine learning and Agri-tech journals.

This project is a collaboration between the ARC Training Centre for Smart & Sustainable Horticulture (TC-SaSH) and University of Adelaide (UoA). The project will be conducted at the Western Sydney University node of the TC-SaSH on the Hawkesbury Campus, University of Adelaide and University of Western Australia. The project is led by Associate Professor Oula Ghannoum & Associate Professor Yi Guo, supported by a team of plant physiologists at the Hawkesbury Institute for the Environment (HIE), Professor Zhonghua Chen at the University of Adelaide and Distinguished Professor Sergey Shabala at University of Western Australia. HIE is a research institute within WSU and has rapidly become a research leader in plant environmental and ecological research, with a strong reputation for delivering research outcomes of the highest quality. This research will uncover the novel germplasm for high-level production inside a protected environment.

What does the scholarship provide?

• Domestic candidates will receive a tax-free stipend of $37,000 (AUD) per annum for up to 3 years to support living costs, supported by the Research Training Program (RTP) Fee Offset.
• International candidates will receive a tax-free stipend of $37,000 (AUD) per annum for up to 3 years to support living costs. Those with a strong track record will be eligible for a tuition fee waiver and an Overseas Student Health Cover (OSHC) Single Policy.
• Support for conference attendance, fieldwork and additional costs as approved by the Institute.

Eligibility Criteria

We welcome applicants from a range of backgrounds, who are keen to apply their skills to key issues in crop physiology in protected facilities. In particular, the project is suitable for candidates with strong interests in photosynthesis and using state-of-the-art equipment to assess parameters that influence carbon assimilation.

The successful applicant should:

• A Bachelor's degree with First Class Honours (or equivalent) or a Master's degree (by research or coursework with a substantial research component) in Computational/applied Mathematics/physics, Data Science, Computer Science, or a closely related quantitative discipline.
• Demonstrated strong programming skills in Python and experience with deep learning frameworks (preferably PyTorch).
• Solid background in computer vision (e.g., CNNs, autoencoders, vision transformers, or image sequence processing) and some prior knowledge in dynamical systems (e.g., ordinary differential equations, control theory, system identification, neural differential equations, or recurrent/sequential modelling of time-evolving processes).
• Experience or strong interest in time-series modelling, sequence prediction, or video understanding (e.g., video prediction, latent dynamics, or conditional generative models) is highly desirable.
• Excellent analytical and problem-solving skills, with the ability to work independently on complex, interdisciplinary problems at the intersection of machine learning and plant science.
• Good written and oral communication skills in English, suitable for publishing research findings and presenting at international conferences.
• Willingness to engage with real-world greenhouse data collection and collaborate with horticultural researchers.

International applicants must demonstrate English language proficiency.

Lead Researcher: Professor Oula Ghannoum

Applications close: 31 May 2026