Research Associate/Fellow (Fixed-term)
- Closing Date
- Tuesday, 6th March 2018
- Job Type
- Computer Science
- £26495 to £38833 per annum (pro-rata if applicable) depending on skills and experience (minimum £29799 with relevant PhD). Salary progression beyond this scale is subject to performance
Project: Deep Learning for Image Analysis
Image analysis is often used to count or measure objects in scientific images. Deep learning has, in the last few years, begun to transform what is possible in the field of image analysis. This is especially true for images containing challenging subject matter. This project will analyse images from just such a challenging subject: plants. There is a desire in plant phenotyping to automatically measure characteristics of plants, crops and related matter such as seeds automatically to quantify, for example, plant responses to environmental stress (e.g. low water, low nutrients), disease, pests, etc. Recently we have found deep learning to be a promising foundation to help measure such parameters in a high throughput manner (e.g. https://academic.oup.com/gigascience/article/doi/10.1093/gigascience/gix083/4091592/Deep-Machine-Learning-provides-state-of-the-art). The motivation for this work is to address food security concerns; to produce the high-performing crops needed to feed the population of 9 billion people expected by 2050.
This 3-year project will adapt, develop and apply deep-learning techniques to a range of plant-phenotyping and formulation chemistry derived data sets, in order to determine how well the approach can perform in challenging environments.These will span from lab-based plate assays, through to glasshouse and field-derived images.We will work closely with the project partner Syngenta (www.syngenta.co.uk), a global agricultural company who will provide the image sets.
The candidate should have a PhD (or be close to completion) in image analysis or a deep learning-related subject. The ability to work in an interdisciplinary team will be essential. The ability to develop new CNN models/architectures, as well as use existing architectures, will be required. The candidate will be based in the School of Computer Science, Jubilee Campus at the University of Nottingham, but will be required to spend some time at Syngenta’s international research centre in Berkshire.
This full-time post is fixed-term for a period of 3 years. Job share arrangements may be considered for this post.
Informal enquiries may be addressed to Andrew French, tel: 0115 951 6374 or email email@example.com. Please note that applications sent directly to this email address will not be accepted.
The University of Nottingham is an equal opportunities employer and welcomes applications from all sections of the community.