Data Scientist - GIS-based building detection
Result of Service Mapping informal settlements in Belmopan, Guatemala City, San Salvador, Tegucigalpa, Managua, San José, Panama City and Santo Domingo and making the shapefiles, including basic spatial data made available in a website.
Work Location home-based
Expected duration 7 months
Duties and Responsibilities 1. About UN-Habitat The United Nations Human Settlements Programme (UN-Habitat) is the lead United Nations agency for cities and human settlements. The agency is mandated by the UN General Assembly to promote socially and environmentally sustainable towns and cities with the goal of providing adequate shelter for all and sustainable development. The main documents outlining the mandate of the organization are the Vancouver Declaration on Human Settlements, the Habitat Agenda, the Istanbul Declaration on Human Settlements, the Declaration on Cities and Other Human Settlements in the New Millennium, and UN General Assembly Resolution A/56/206. UN-Habitat, being the focal point for all urbanization and human settlement matters within the UN system, has a role in delivering the 2030 Sustainable Development Agenda, adopted by Member States in 2015, specifically goal 11: make cities and human settlements inclusive, safe, resilient and sustainable. As our cities and towns grow at unprecedented rates, setting the social, political, cultural and environmental trends of the world, sustainable urbanization is one of the most pressing challenges for the global community in the 21st century. In 1950, one-third of the world’s population lived in cities. Just 50 years later, this proportion has risen to one-half and will continue to grow to two-thirds, or six billion people, by 2050. Cities are now home to half of humankind. They are the hub for national production and consumption - economic processes that generate wealth and opportunity. But they also create disease, crime, pollution and poverty. In many cities, especially in developing countries, slum dwellers number more than 50% of the population and have little or no access to shelter, water and sanitation.
2. United Nations Innovation Technology Accelerator for Cities (UNITAC Hamburg) The United Nations Innovation Technology Accelerator for Cities is an innovation lab in Hamburg, established in March 2021 by UN-Habitat in collaboration with the UN Office of Information and Communications Technology (UN-OICT) and HafenCity University. The Accelerator promotes open and participatory governance of data and digital platforms, innovations related to mapping, spatial analysis and data visualization and people-centred smart cities. UNITAC Hamburg is funded by the Government of Germany for four years. The Accelerator is creating a team of urban, technology and innovation experts which will develop technologies to solve the world’s most pressing urban challenges and ensure that we speed up progress towards achieving the Sustainable Development Goals, particularly Goal 11 on sustainable cities and communities. UNITAC Hamburg works with a bespoke urban technology innovation acceleration methodology that can be used to analyse real urban challenges and urbanization trends. The methodology will be based around three themes: (1) Open, transparent and participatory governance of data and digital platforms (2) Mapping spatial analysis and visualization, and (3) People-centred smart cities. Within these fields, particular emphasis is placed on how to generate knowledge and use cases on smart city governance and the development of frontier technologies with a special emphasis on developing countries and cities with large informal populations. The solutions developed by UNITAC Hamburg will be tested and piloted through UN-Habitat’s networks across the world with a particular focus on less developed countries. UNITAC Hamburg will also work with the City of Hamburg to establish an urban technology acceleration testbed in Hamburg in which technologies, governance frameworks and approaches can be tested in a simulated urban environment before scaling up internationally. Learning will be documented and published through high-quality guides and toolkits that can be used by policy makers to ensure that smart cities are inclusive, sustainable and people-centered.
3. Objective BEAM is a specialized software, developed by UNITAC, for the AI/ML-based detection of buildings in informal settlements, using aerial images, to support local governments in developing strategies, identifying areas for priority interventions and ultimately to improving the lives of citizens in informal areas.
The objective is to enhance the BEAM tool, to use satellite images and upgrading the building detection logic for mapping informal settlements in 8 regions of intervention (Central America).
The results of the mapping (shape files) shall be made available in an online ‘informal settlements registry’.
DUTIES AND RESPONSIBILITIES
4. Scope of Work: For the UNITAC deliverables, the consultant will work under the overall leadership of the UNITAC Hamburg Programme Management Officer, and closely coordinate with the UNITAC Hamburg team and partner entities. - Upgrading BEAM to use satellite images (UNOSAT or similar) - Developing AI/ML based logic for detecting individual buildings in informal setting with an acceptable level of accuracy. - Optimizing the application for time- and resource-efficiency. - Mapping informal settlements in 8 Central American capitals (Belize, Belmopan; Guatemala, Guatemala City; El Salvador, San Salvador; Honduras, Tegucigalpa; Nicaragua, Managua; Costa Rica, San José; Panama, Panama City; Santo Domingo, Santo Domingo). - Providing a spatial overview of number, location, extent and the principal morphological characteristics of informal settlements. - Adaptation of the image detection logic for 8 different regions of interventions (cities with 1 to 1.5 million inhabitants). - Creation of ‘proof of concept’, including workplan preparation, the identification of data sources, setting up of pipeline and creation of sample training data and model - Creation of additional training data for model refinement and improvement - Model development, including the creation of final training data and technical augmentations, final model training and testing - Technical briefing and training on the use of the AI tool and how models were trained - Provide handover of model and model output (creation of shapefiles) - Creating technical documentation including description of logic used, source code documentation, installation and user manual.
5. Additional Requirements: - Development, programming, testing, debugging and implementation of BEAM enhancements, using open-source technologies, open data and data provided as part of the project (satellite images), - Work as part of a multidisciplinary team and provide guidance / advice on the use of GIS, AI and ML. - Work with stakeholders to develop upgrades that meet their requirements. - Perform other duties as assigned.
Qualifications/special skills Education: Advanced university degree (Master’s degree or equivalent) in Computer Science, Software Engineering, Geoinformatics or a directly related field or its equivalent in education is required. A first-level university degree in combination with two (2) years’ experience may be accepted in lieu of the advanced university degree.
Work Experience: - 10 years' experience with applied statistics, machine learning algorithms, classification, prediction, and deep learning methods is required. - Experience with image recognition and segmentation - Experience on Convolutional Neural Networks method in addition to other classification and prediction machine learning techniques - Experience conceptualizing, implementing, and maintaining Machine Learning models especially linked to satellite or arial imagery - Deep knowledge about U-net and HR-net models - Advanced programming skills in Python (other programming languages such as R, JavaScript, C++, etc.) would be an advantage - Experience with geospatial data e.g.QGIS tool and shapefiles
Languages English and French are the working languages of the United Nations Secretariat. For this position, fluency in English is required. Knowledge of German is desirable, and another UN language is desirable.
No Fee THE UNITED NATIONS DOES NOT CHARGE A FEE AT ANY STAGE OF THE RECRUITMENT PROCESS (APPLICATION, INTERVIEW MEETING, PROCESSING, OR TRAINING). THE UNITED NATIONS DOES NOT CONCERN ITSELF WITH INFORMATION ON APPLICANTS’ BANK ACCOUNTS.