Automatic Number Plate Recognition
Experimental results show that the identification accuracy of the proposed ANPR exceeds 95%, demonstrating its potential in practical applications and suggestions for future research to improve stability and efficiency.
Abstract
This final project develops an algorithm for automatic number plate recognition (ANPR). ANPR has gained much interest during the last decade along with the improvement of digital cameras and the gain in computational capacity. \n\t\t\t\t The text is divided in four chapters. The first, introduces the origins of digital \n\t\t\t\t image processing, also a little resume about the following algorithms that are \n\t\t\t\t needed for develop the system ANPR. The second chapter presents the objectives \n\t\t\t\t to be achieved in this project, as well as, the program used for his development. \n\t\t\t\t The following chapter explains the different algorithms that compound the system, \n\t\t\t\t which is built in five sections; the first is the initial detection of a possible number \n\t\t\t\t plate using edge and intensity detection to extract information from the image. The \n\t\t\t\t second and the third step, thresholding and normalization, are necessary to use the \n\t\t\t\t images in the following stages; the text of the plate is found and normalized. With \n\t\t\t\t the segmentation, each character of the plate is isolated for subsequent recognition. The last step reads the characters by correlation template matching, which is a simple but robust way of recognizing structured text with a small set of characters. It is evaluated the system’s speed and his error rate. Finally, the conclusions and future works are shown in the chapter four. The databases used consist of images under normal conditions and only Bulgarian’s numbers plate.