Research on License Plate Recognition System Based on Computer Vision
In order to achieve the exact location and character recognition of license plate, firstly, this paper got binary image of license plate and done edge detection with differential operation. Secondly, it searched the license plate binary image after difference for the horizontal and vertical cut point, and determined the best cutting threshold through the experiment. Finally, it made the character segmentation by vertical projection, the recognition of license plate characters with the use of BP neural network, whose overall recognition rate is at 95.3%, and the display interface design for program transfer and results display. The experimental results showed that the location of license plate was exact and the character recognition rate was high.
Zhenyu Du and Bin Liu
Y. Q. Wang et al., "Research on License Plate Recognition System Based on Computer Vision", Applied Mechanics and Materials, Vol. 65, pp. 536-541, 2011