{"id":32296,"date":"2024-07-08T09:56:41","date_gmt":"2024-07-08T07:56:41","guid":{"rendered":"https:\/\/nr.no\/en\/?post_type=bc_area&p=32296"},"modified":"2025-10-01T14:27:37","modified_gmt":"2025-10-01T12:27:37","slug":"deep-learning-for-complex-image-data","status":"publish","type":"bc_area","link":"https:\/\/nr.no\/en\/areas\/deep-learning-for-complex-image-data\/","title":{"rendered":"Deep learning for complex image data"},"content":{"rendered":"\n
\n
\n

Deep learning (DL) has revolutionised computer vision and significantly transformed the fields of image analysis and machine learning. We have been at the forefront of adapting DL for numerous applications and it has been integral to most of our projects in image analysis since 2018. <\/strong><\/p>\n\n\n\n

NR contributes to research and development in image analysis, machine learning, and Earth observation. Our applications are diverse, spanning healthcare, transportation, ocean, climate and environment, infrastructure, and general mapping and monitoring. We use deep learning to analyse complex image data from a range of sources, including ultrasound sensors, MRI machines, echo sounders, seismic sensors, multi-spectral and hyperspectral sensors, and synthetic aperture radar (SAR) satellite sensors. <\/p>\n\n\n\t

\n\t<\/div>\n\t\n\n\n
\n
\n

Deep learning exceeds human performance<\/h2>\n\n\n\n

Today, deep learning is the leading approach for solving most image analysis problems, often achieving accuracies that surpass human performance. We have successfully applied deep learning to a wide range of real-world applications, with many of our solutions in operational use every day.<\/p>\n<\/div>\n<\/div>\n\n\n\n

<\/p>\n\n\n\n

\"Three
Deep learning can be applied to solve a variety of urgent problems. From left to right: 1. Interpreting seismic data. Image: Equinor\/NR.  2. Daily mapping of snow cover in Norway using satellite images. Image: ESA\/NR. 3. Detecting cancerous growth in breast cancer screening images. Image: The Cancer Registry of Norway\/NR.<\/em><\/figcaption><\/figure>\n\n\n\n

Research challenges<\/strong><\/h3>\n\n\n\n

Learning and applying deep learning comes with several challenges. Annotating complex image data requires expert knowledge and is both costly and time-consuming. While machine learning methods are data-driven, incorporating human knowledge, constraints and physics is often necessary. To ensure trustworthiness, it is crucial that the models acknowledge their own limitations. Additionally, due to their complexity, the models often lack clarity and transparency during interference.<\/p>\n\n\n\n

Our research encompasses the following areas:<\/p>\n\n\n\n