Title :
An intelligent framework of illumination effects elimination for Car License Plate character segmentation
Author_Institution :
Dept. of Comput. Eng., Sejong Univ., Seoul, South Korea
Abstract :
In computer vision, Automatic Car License Plate Recognition is popular research area. Many methods for Car License Plate Recognition has been developed, however, a car license plate which is degraded by illumination or dirt effects may yield false recognition because degradation elements interfere character segmentation. Although many researches for reducing degradation effects on a car license plate are established, research for degradation of various illumination effects is insufficient. This paper introduces an intelligent framework that outlines character of car license plate which is degraded by various illumination effects. Our framework shows robustness for outlining character of car license plate image under various lightning or illumination effects.
Keywords :
character recognition; computer vision; image segmentation; lighting; automatic car license plate recognition; car license plate character segmentation; car license plate image; character segmentation; computer vision; degradation element; illumination effects elimination; robustness; Artificial neural networks; Cybernetics; Degradation; Image edge detection; Image segmentation; Licenses; Lighting; Binarization; Edge detection; Illumination; License plate character segmentation; Machine learning; Neural network;
Conference_Titel :
Machine Learning and Cybernetics (ICMLC), 2010 International Conference on
Conference_Location :
Qingdao
Print_ISBN :
978-1-4244-6526-2
DOI :
10.1109/ICMLC.2010.5580887