Data Modelling Technologist (DMT)

Organization

iMMAP is an international nongovernmental organisation that provides information management services to humanitarian and development organizations. Through information management, we help our partners target assistance to the world’s most vulnerable populations. Our core philosophy is that better data leads to better decisions and that better decisions lead to better outcomes. iMMAP’s critical support to information value chains helps to solve operational and strategic challenges of our partners in both emergency and development contexts by enabling evidence-based decision-making for better outcomes.

Background

The Data Modelling Technologist (DMT) position is a short-term service contract position that is required to assist the iMMAP GeoInformatics Unit to implement and test next generation change detection algorithms on the latest Cubesat data from Planet Labs. The aim of the work is to assess the potential of Planet satellite data as a future data source in the Open Data Cube (ODC), with the focus on change detection in urban environments affected by conflict. Enhanced change detection algorithms will be applied to Planet data for specific cities in the MENA region, which has been affected in various degrees by conflict over the last few years. The project will attempt to identify significant changes within the urban environment between 2018 – 2020, and though validation with ground truth information attempt to classify the type of changes observed.

The majority of change detection approaches utilize an image to image comparison that is based on image differencing or the ratio between images. The availability of imagery at a much higher revisit rate has opened the door for more sophisticated approaches to be utilized by viewing the imagery set as a time-series rather than a sparse set of individual images. This approach has been proven very useful in coarse resolution imagery where, for example, daily imagery has been available globally for some time. The proposed change detection approach will evaluate the change in distribution (using a density similarity index such as F-divergence or Bhattacharyya distance) over a sliding window of the image (i.e. a small area of the image will be considered over all the time-steps) and the intensity distribution will be calculated at each point in time. The change in the distribution over time will be used as a proxy for change, the sub window will be shifted to the full extent of the image in making the method inherently robust to co-registration errors.

In this regard, he/she will manage relationships with the lead organizations, in preparing and developing high quality objective plans, concepts, processes, report to monitoring and evaluation offices, and enhance internal and external representation of the objectives.

Description of Duties KEY RESPONSIBILITIES

The DMT (in line with the wider iMMAP Strategy) is responsible for the following:

• Assist in the design, development and implementation of key change detection algorithms

• Preparation of key Planet labs data

• Oversee project parameters and day-to-day implementation of the project

• Application and testing of time-series based change detection algorithms for the selected cities

• Evaluation of change detection outputs in relation to ground truth data within cities. Damage assessment reports as well as VHR data will be utilized to assess the sensitivity to different types of change

• Produce a change detection product for each selected city identifying: Minimal, Medium and Significant change locations, 2018 – 2020

With specific relation to the ODC platforms:

• Coordinate the System concept design and Pilot with the Project Manager and Lead Developer

• Finalize the Development including documentation manual and code on GitHub in English for users and developers.

Any other duties as required.

Requirements

REQUIRED

• A degree in a relevant discipline /or equivalent experience

• A minimum of 5 years progressively responsible experience in a similar role

• Experience in Remote Sensing image processing and data analytics

• Expert Python programming skills

• Expert in Data modelling and change detection algorithms

• Great communication skills WORKING CONDITIONS

• Willingness to work under tight deadlines.

• Supervision of a diverse team of individuals.

• Physically fit and able to cope with demands of unusual working hours and difficult environments.

Personnel is expected to conduct itself in a professional manner and respect local laws, customs and iMMAP’s policies, procedures, and values at all times and in all Countries.

iMMAP has zero-tolerance for sexual harassment, exploitation and abuse, human trafficking, child abuse and exploitation. Any violations of these principles and policies will be treated as serious misconduct.

iMMAP is an Equal Opportunity Employer regardless of background.