Title :
Automatic license extraction from moving vehicles
Author :
Cui, Yuntao ; Huang, Qian
Author_Institution :
Siemens Corp. Res. Inc., Princeton, NJ, USA
Abstract :
We present a new approach to extract the license from an image sequence of moving vehicles. The approach includes the following components: 1) license plate localization; 2) feature extraction and tracking; 3) perspective distortion correction; 4) binarization. We model the binarization of characters as a Markov random field (MRF), where the randomness is used to describe the uncertainty in pixel label assignment. With the MRF modeling, the extraction of characters is formulated as the problem of maximizing the a posteriori probability based on given prior and observations. A genetic algorithm with local greedy mutation operator is employed to optimize the objective function based on MRF modeling. In the experiments, we compared our results with other two methods that were evaluated. Our method has demonstrated better performance
Keywords :
Markov processes; feature extraction; image classification; image sequences; optical character recognition; optical tracking; road vehicles; MRF modeling; Markov random field; a posteriori probability; automatic license extraction; binarization; feature extraction; feature tracking; genetic algorithm; image sequence; license plate localization; local greedy mutation operator; moving vehicles; objective function; performance; perspective distortion correction; pixel label assignment; Character recognition; Computerized monitoring; Educational institutions; Feature extraction; Image recognition; Licenses; Optical character recognition software; Optical distortion; Road vehicles; Vehicle detection;
Conference_Titel :
Image Processing, 1997. Proceedings., International Conference on
Conference_Location :
Santa Barbara, CA
Print_ISBN :
0-8186-8183-7
DOI :
10.1109/ICIP.1997.632014