Image analysis and Earth observation
- Department BAMJO
- Fields involved Machine learning, Earth observation, Image analysis, Climate and Environment, Mapping and surveillance, Monitoring sea and lake ice
- Industries involved Climate and environment, Health, Technology and industry, Ocean, Natural resources, Community
With close to 40 years of experience, NR has extensive knowledge in image analysis, Earth observation and machine learning, and practical experience with a wide range of applications. Detection, characterisation and object recognition are central themes in many of our projects.
In addition to object detection, characterisation, recognition and extraction in images and image sequences, we also work with data sequence analysis and pattern recognition of unstructured text. Notably, pattern recognition is not limited to images, but can also be applied to other types of data, such as audio, accoustic signals and written text. Our success often lies in simple, yet efficient solutions using approaches that combine preestablished knowledge and known context with the observed data presented.
Image analysis
Within general image analysis we work in areas like medicine and healthcare, marine, industry and energy.
Earth observation
Earth observation analyses digital remote sensing data from satellites and aircraft. Furthermore, we develop methods, algorithms and tools for object recognition and classification, as well as size computations based on physical modelling.
Deep learning
Deep learning, a subset of machine learning, focuses on artificial neural networks and their ability to process and make decisions similarly to the human brain.
Deep learning has drastically surpassed previous classification methods in terms of accuracy, and it is central in speech recognition, visual object recognition, character recognition, and several language related tasks.
Deeper machine learning architectures are better at handling complex recognition tasks compared to earlier, shallower models.
Benefits include:
- superior modelling capabilities of heterogeneous data in layers of increasing complexity
- the ability to learn the best features to represent raw data
- the ability to improve performance as more training data is made available
Earth observation, surveillance and climate
Image analysis and machine learning
Research areas
Selected projects
Selected research topics
We analyse images taken from drones, trains and aircraft with deep neural networks that can detect infrastructural faults and maintenance requirements automatically. Recent advancements in automation have transformed inspection processes, reducing expenses, creating safer work environments for employees, and greatly improving the quality of infrastructure in Norway and abroad.
We have developed methods for automated monitoring of sea and lake ice to assess properties such as surface temperature and ice thickness. Monitoring is important for navigation purposes, climate change observation and public safety.