Automated optic recognition of alphanumeric content in car license plates in a semi-structured environment

Abstract

Optical character recognition (OCR) encompasses a large number of applications, which have appeared in the scientific literature. The spectrum of applications ranges from simple character recognition of printed sets to advanced handwritten text recognition, and has yielded satisfactory results in many cases. The present work focuses on the recognition of the alphanumeric content in car license plates within a semi-structured environment, using the adept vision system. A systematic approach, addressing extensively the associated theoretical and practical issues, was developed and is presented. The approach utilizes the adept assembly and information management environment to integrate in a single system the acquisition and digital image processing stages with the system training and the results interpretation. Several tests were conducted in order to evaluate the performance of the system and to analyze its sensitivity. The obtained results demonstrate the robustness of the system and the feasibility of operation in outdoor environments.