Computer Vision
Computer vision is concerned with the automated analysis of still images and videos. We analyse images created by standard digital color cameras, active depth sensors (e.g. Microsoft Kinect), as well as multiple band images from astronomical telescopes.
Computer vision is rapidly becoming an essential ingredient in many daily life activities such as automatic driving assistants, face and fingerprint recognition, natural user interfaces for computers and game consoles, mobile phone applications as well as applications involving consumer digital cameras or video surveillance.
We focus on development of fundamental algorithms as well as applications of these to industrial and real-life problems, such as welding, satellite imagery, food production, saw mill industry, agriculture, human rehabilitation as well as astronomical image analysis.
Computer vision and image analysis are closely related to neighboring research areas such as machine learning, signal processing and graphics, and our research is mathematically founded combined with empirical evaluation.
- Using UAV drones for mapping weeds in fields of crop (Drone)
- Texture- and color-based segmentation of weeds in crop
- Surveying the sky using machine learning (SkyML)
- Applications of computer vision methods such as texture analysis to astronomical data
- TreeDFurniture
Automatic defect detection from images of high quality furniture wood - Image features (interest point detectors and descriptor)
- Comparative study of interest point detectors and descriptors and development of novel algorithms
- Human motion imitation (HUMIM)
- Modelling, tracking and analysis of human motion, computer-assisted physiotherapeutic rehabilitation
- Real-time Robots in the Meat industry
- Real-time welding control2D shape modelling, image processing
People
| Name | Title | Phone | |
|---|---|---|---|
| Francois Lauze | Associate Professor | +4535335671 | |
| Jon Sporring | Professor | ||
| Kim Steenstrup Pedersen | Professor | +4535321455 |