Image processing
EMVA presents Young Professional Award 2018
The Young Professional Award 2018 of the EMVA (European Machine Vision Association) goes to Doris Antensteiner for her work 'Light Field and Photometric Stereo'.
The reconstruction of the 3D surface and reflectance properties of objects is a fundamental problem in image processing. Light field is usually defined as a plenoptic 4D function that contains ray information of light and consists of two spatial and two directional dimensions. The light field cameras currently available on the market include plenoptic cameras, matrix cameras and industrial multi-line scan cameras. In Doris Antensteiner's work, light field information is used to create a 3D surface reconstruction and to record the reflectance and material properties of objects.
In the work awarded by the EMVA, light rays reflected by objects are analyzed under defined illumination conditions. Conventional imaging methods do not allow the recording of directional information of the incident light and record the sum of the incident light. Therefore, only a 2D representation of the scene is created. The work pushes for a more comprehensive scene description using light field and photometric stereo. An optimal fusion of light field with photometric stereo is achieved using variation methods for both area scan and multi-line scan cameras. The latter are mostly used in industrial in-line environments where objects move on conveyor belts at constant speed. The work presents solutions for the combined 3D surface reconstruction of depth data with surface normals in industrial in-line environments.
Applications of these methods include product inspection, defect detection, brand protection, product safety and optical inspection of materials. High-precision 3D reconstructions allow detailed defect detection in industrial inspection.
Doris Antensteiner (32) received a master's degree in computer science from the Vienna University of Technology in 2011 and a further master's degree in computer science management in 2014. From 2012 to 2015, she worked as an 'Image Processing Engineer' in the 'R&D Video and Sensors' department at Kapsch TrafficCom. She is currently working at the Austrian Institute of Technology Center for Vision, Automation and Control and is doing her doctorate at Graz University of Technology.










