Translator Disclaimer
Paper
17 August 1994 Identification of simple objects in image sequences
Author Affiliations +
Proceedings Volume 2357, ISPRS Commission III Symposium: Spatial Information from Digital Photogrammetry and Computer Vision; (1994) https://doi.org/10.1117/12.182807
Event: Spatial Information from Digital Photogrammetry and Computer Vision: ISPRS Commission III Symposium, 1994, Munich, Germany
Abstract
We present an investigation in the identification and location of simple objects in color image sequences. As an example the identification of traffic signs is discussed. Three aspects are of special interest. First regions have to be detected which may contain the object. The separation of those regions from the background can be based on color, motion, and contours. In the experiments all three possibilities are investigated. The second aspect focuses on the extraction of suitable features for the identification of the objects. For that purpose the border line of the region of interest is used. For planar objects a sufficient approximation of perspective projection is affine mapping. In consequence, it is near at hand to extract affine-invariant features from the border line. The investigation includes invariant features based on Fourier descriptors and moments. Finally, the object is identified by maximum likelihood classification. In the experiments all three basic object types are correctly identified. The probabilities for misclassification have been found to be below 1%
© (1994) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Christoph Geiselmann and Michael Hahn "Identification of simple objects in image sequences", Proc. SPIE 2357, ISPRS Commission III Symposium: Spatial Information from Digital Photogrammetry and Computer Vision, (17 August 1994); https://doi.org/10.1117/12.182807
PROCEEDINGS
8 PAGES


SHARE
Advertisement
Advertisement
RELATED CONTENT


Back to Top