Master of Engineering scholarships


Do you want to help develop the latest technology for biosecurity surveillance in Aotearoa?

We have four positions available with our State-of-the-art surveillance team, which will focus on providing ‘proof of concept’ for image analysis in biosecurity.

The projects will develop imaging acquisition, data storage and processing as well as computational intelligence that can be used to detect changes in plant health in urban vegetation as an early indicator of the presence of unwanted pests or diseases of biosecurity concern.

The project aims to create an efficient large area coverage image-based surveillance system that applies automation in knowledge intensive processes such as data acquisition, image processing, stress and disease recognition and anomaly detection. This enables prioritisation of human inspection of suspect symptoms to improve the early detection of invasive pests.

Each scholarship is worth NZ$30,000 stipend + fees and are available at either the University of Canterbury (Computer Science or Mathematics and Statistics departments) or the University of Waikato (School of Engineering).


Please note each position has different start dates and institutions. While any person listed can discuss the projects in general, make sure to send your CV, transcripts and a statement of interest to the relevant supervisor.


Master Research Project 1: Urban Vegetation Imaging Surveillance System

Location: University of Waikato
Start date:
November 2021
Supervisors:
Dr. Ye Chow Kuang, Prof. Mike Duke, and A/Prof. Melanie Ooi with co-supervison by University of Canterbury collaborators Prof. Richard Green, Dr. Varvara Vetrova and Dr. Steve Pawson.

Send CV, transcripts and a statement of interest to: ykuang@waikato.ac.nz

Design and develop specification for the sensing, data communication, data storage system. The system is expected to attach easily to vehicles, collect high quality images of urban vegetation while moving at urban speed. In addition, the housing should operate reliably under various New Zealand weather conditions and rough handling expected of urban outdoor environments. Regular on-road system testing and data collection at the identified sites is part of the duty.

Required skill: Relevant BEng(hon) or equivalent qualification. Mechanical/mechatronics CAD design and manufacturing, system integration, digital image processing and analysis, working knowledge of machine learning, New Zealand class 1 driver licence.


Masters Research Project 2: Deep learning-based methods for fine-grain recognise of anomalies in trees

Location: University of Canterbury
Start: November 2021
Supervisors: Dr. Varvara Vetrova, Prof. Richard Green, and Dr. Steve Pawson with co-supervison by University of Waikato collaborators, Dr. Ye Chow Kuang, Prof. Mike Duke, and A/Prof. Melanie Ooi

Send CV, transcripts and a statement of interest to: varvara.vetrova@canterbury.ac.nz

This MSc project is aimed at investigating, designing, and developing models based on deep neural networks that are capable of identifying tree/plant species and detecting anomalies based on photographic images.

Required skills: Bachelor with Honours in a computational/mathematical/software engineering discipline, machine learning background is strongly preferable, strong analytical skills, proficiency in programming languages (Python, Pytorch/Tensorflow libraries).


Master Research Project 3: Digital twin and data fusion for vegetation monitoring

Location: University of Waikato
Start: April 2022
Supervisors: Dr. Ye Chow Kuang, Prof. Mike Duke and A/Prof. Melanie Ooi with co-supervision by University of Canterbury collaborators, Prof. Richard Green, Dr Varvara Vetrova and Dr. Steve Pawson.

Send CV, transcripts and a statement of interest to: ykuang@waikato.ac.nz

Design a software platform that integrates information, meteorological data augmentation and facilitate model building to convert information into knowledge. Tasks include privacy blurring, building digital twin from the urban vegetation data. Combine GIS data and plan species recognition algorithms to detect anomalies.

Required skill: BSc(Hon) or equivalent in computational/software engineering discipline. Proficiency in programming, working knowledge on GIS, computer vision and machine learning.


Master Research Project 4: Deep learning-based methods for detecting known tree diseases

Location: University of Canterbury
Start: April 2022
Supervisors: Prof. Richard Green, Dr. Varvara Vetrova, and Dr. Steve Pawson with co-supervison by University of Waikato collaborators, Dr. Ye Chow Kuang, Prof. Mike Duke, and A/Prof. Melanie Ooi

Send CV, transcripts and a statement of interest to: Richard.Green@canterbury.ac.nz

The focus of this project is to develop a proof of concept deep learning approach to identify known biosecurity threats. This will require integration of known tree species information (from Master’s Project 2) to detect symptoms of known diseases that affect these host plants.

Required skill: BSc(Hon) or BE in computer science or software engineering (or equivalent) with a working knowledge of deep learning and computer vision.

back to top