Job Board

Job Description

We are seeking a highly qualified postdoctoral researcher to join the Map Automation research group at the University of Winnipeg.

The successful candidate will work on a project to develop an approach for automated map production from multispectral satellite images using deep learning methods. Specifically, convolutional neural networks designed for classifying each pixel in a satellite image will be used to produce linear disturbance maps from Sentinel-2 data. The main research challenge to be addressed is that labels were developed using commercial high-resolution SPOT-6 satellite data, and there are no equivalent labels for the free and medium-resolution Sentinel-2 data. Thus, the proposed work will develop an unsupervised domain adaptation approach, which aims to take the data and corresponding labels from one domain and adapt it for training semantic segmentation models in a related, but different domain. The position also includes supervisory duties of students working at the BSc, and MSc level.

The context of the work for this position is linear disturbance mapping. In Canada’s vast northern region of boreal forest and wetlands, disturbances such as roads, seismic exploration, pipelines, and energy transmission corridors are a leading cause of the decline of woodland caribou (Rangifer tarandus) – boreal population. As a result, a deep understanding of these “linear disturbances” has become a research and forest management priority in Canada. An ideal tool to support managing linear disturbances is a way to automatically generate cost-effective maps that accurately identify this form of forest habitat fragmentation.

This position is part an industry-academia collaboration between the University of Winnipeg and Hatfield Consultants LLP. The successful candidate will work and interact with Hatfield remote sensing specialists, and the successful candidate will spend time in both Winnipeg and Vancouver.

Hatfield is a privately-owned, multidisciplinary company established in 1974 with over 4,000 projects successfully completed in over 40 countries. Our wide range of consulting services include environmental and social impact assessment, environmental management and monitoring of infrastructure projects, geomatics and remote sensing, and environmental information systems. Hatfield’s head office is located in North Vancouver, British Columbia (BC), with regional BC offices in Fort St. John, Terrace, Vernon, and New Westminster, and Fort McMurray and Calgary, Alberta.

Responsibilities:
1. Work in a collaborative team environment, under the general direction of Drs. Christopher Henry and Christopher Storie to achieve the goals of associated research grant funding requirements;
2. Conduct high-level academic literature reviews and associated writing tasks;
Analyze and pre-process remote sensing satellite data for using in machine learning models;
Design, train, and test deep learning neural network architectures and associated models;
Draft technical reports
3. Contribute to academic and popular publications related to project research;
4. Other duties, as assigned.

Required Qualifications:
1. The candidate will possess a Ph.D. in computer science, engineering, mathematics, statistics, physics, or related fields
1a. Will consider a candidate with an MSc in above fields for appointment as a Research Associate/Assistant
2. Excellent oral and written communication skills
3. Excellent organizational skills with the ability to manage time effectively and efficiently to meet deadlines
4. Strong interpersonal skills with the ability to work as part of a team

Desired Qualifications (one or more of the following):
1. Formal training in machine learning, especially convolutional neural networks, generative adversarial networks, and domain adaptation networks
2. Experience in developing machine learning applications
3. Experience working with digital image and remote sensing data
4. Experience working with large geospatial datasets
5. Knowledge and experience with Python, and TensorFlow or PyTorch
6. Knowledge and experience with Docker
7. Knowledge and experience with Linux

Position subject to budgetary approval, and position wage or salary will be commensurate with experience and education.

Condition(s) of Employment:
*Must be legally entitled to work in Canada.

The University of Winnipeg has a Mandatory COVID-19 Vaccination Policy that requires all those coming to designated indoor campus spaces to provide proof they are fully vaccinated. The Policy was suspended effective May 2, 2022. However, prospective employees should know that the Policy may be reinstated in the future based on changing public health circumstances.

The University of Winnipeg is committed to equity, diversity and inclusion and recognizes that a diverse staff and faculty benefits and enriches the work, learning and research environments, and is essential to academic and institutional excellence. We welcome applications from all qualified individuals and encourage women, racialized persons, Indigenous persons, persons with disabilities, and 2SLGBTQ+ persons to confidentially self-identify at time of application.

The University of Winnipeg is committed to ensuring employment opportunities are accessible for all applicants. If you require accommodation supports during the recruitment process, please contact [email protected]

The personal information of applicants is collected under the authority of the University of Winnipeg Act and 36(1)(b) of the Freedom of Information and Protection of Privacy Act. All personal information collected via the recruitment process is used to assess the applicant’s suitability, eligibility, and qualifications for employment, and to otherwise support recruitment activities. This information will be provided to participating members of the recruitment process. Questions regarding the collection of your personal information may be directed to the Director, HR Services, 515 Portage Avenue, Winnipeg, MB, R3B 2E9 or 204.786.9066.

All applicants must apply via this link – https://www.northstarats.com/University-of-Winnipeg/Postdoctoral-Fellow-Exploring-Unsupervised-Domain-Adaptation-Methods-for-Automated-Linear-Disturbance-Mapping/75015 – applications submitted outside this system cannot be considered