Dynamic Vision for License Plate Recognition

The problems related to license plate recognition and management are much more complex than simple character reading of the car-plate on a sharply illuminated, well contrasted, in-focus, still image of a car (front or rear views). There are many system-engineering issues to be addressed when dealing with license plate recognition and tracking. Image acquisition is the first fundamental step of the process, quite critical in such an almost unconstrained environment like outdoor scenes. It requires multidisciplinary competence of electro-optics, and installation constraints. Infrared lighting is widely used especially at night, which minimizes external uncontrolled conditions. The position of the sensor and its orientation are other essential issues, to minimize perspective deformations of the characters and achieve the necessary resolution in the scene. In general, whatever kind of camera is used, it should be placed to avoid occlusion of the license plate, and allow the acquisition of the plate field as much as squared as possible. Angled views of up to 30°, in the horizontal or vertical plane are usually manageable. It is good practice to ensure that the license-plate characters appear as near vertical as possible in the digitized image.

[1]  Anna Maria Colla,et al.  "STRETCH": a system for document storage and retrieval by content , 1999, Proceedings. Tenth International Workshop on Database and Expert Systems Applications. DEXA 99.

[2]  Erik L. Dagless,et al.  VISION BASED ROAD TRAFFIC DATA COLLECTION , 1993 .

[3]  J. Barroso,et al.  Number plate reading using computer vision , 1997, ISIE '97 Proceeding of the IEEE International Symposium on Industrial Electronics.

[4]  Rangachar Kasturi,et al.  Machine vision , 1995 .

[5]  Paolo Ferragina,et al.  Optical recognition of motor vehicle license plates , 1995 .

[6]  Qian Huang,et al.  Extracting characters of license plates from video sequences , 1998, Machine Vision and Applications.

[7]  C. Scagliola,et al.  Generalised projections: a tool for cursive handwriting normalisation , 1999, Proceedings of the Fifth International Conference on Document Analysis and Recognition. ICDAR '99 (Cat. No.PR00318).