Outdoor scene classification is challenging due to irregular geometry, uncontrolled illumination, and noisy reflectance distributions. This paper discusses a Bayesian approach to classifying a color image of an outdoor scene. A likelihood model factors in the physics of the image formation process, sensor noise distribution, and prior distributions over geometry, material types, and illuminant spectrum parameters. These prior distributions are learned through a training process that uses color observations of planar scene patches over time. An iterative linear algorithm estimates the maximum likelihood reflectance, spectrum, geometry, and object class labels for a new image. Experiments on images taken by outdoor surveillance cameras classify known material types and shadow regions correctly, and flag as outliers material types that were not seen previously.
Bayesian color constancy for outdoor object recognition
01.01.2001
864748 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Bayesian Color Constancy for Outdoor Object Recognition
British Library Conference Proceedings | 2001
|British Library Conference Proceedings | 1999
|Bayesian Decision Theory, the Maximum Local Mass Estimate, and Color Constancy
British Library Conference Proceedings | 1995
|