نظام كشف وتمييز لوحة الرخصة للسيارات العراقية == License Plate Detection And Recognition System For Iraqi Cars
Author name:
ثائر عذار هاشم
Supervisor name:
اياد ابراهيم عباس
General topic:
Electrical, Electronic and Communications Engineering
Specific topic:
Electronic Engineering
Degree:
Master
University:
University of Technology - Electrical Engineering Department - Electronic Engineering Branch
Language:
English
University location:
Baghdad
First pages:
34T467 - p.pdf
Abstract:
نظام كشف وتمييز لوحات ارقام السيارات هو تقنية في معالجة الصور يستخدم للتعرف على السيارات من خلال لوحات الارقم الخاصة بها, او بكلمات اخرى هي تقنية تستخدم لتحويل نص صورة لوحة ارقام السيارات الرقمية الى ملف نص ممكن ان يحرر ويستخدم في اي برنامج او تطبيق اخر | Automatic License Plate Detection and Recognition (ALPDR) system is an image processing technology used to identify vehicles by their license plates, or in other words, is a technology used to convert the text on digital images of vehicle license plates to a text file that can be edited and used as such by any other program or application that needs it.License Plate Detection and Recognition system for Iraqi cars have a wide range of applications; it is used to extract plate number to create automated solutions for various problems. Among these applications : parking, border crossing, traffic control, access control and stolen cars tracking.This system consists of two main parts : the first is a practical implementation of how to take a picture automatically for cars passes, this was done by two sensors type (GP2Y0A21YK0F sensor) connected with each other and interfaced with a camera (A4tech USB camera), this camera is interfaced using Matlab with the PC. The camera is attached using USB port. The second part is image processing, this part includes four basic stages : The first stage is image preprocessing which involves image normalization and RGB to gray image conversion. Second stage is detection of a possible license plate using edge detection technique and extracts these LP using region growing technique, detection rates reach to 95%. Third stage is alphanumeric character segmentation to isolate each character, numbers and words of the license plate using Otsu's and Hough transforms technique for subsequent recognition. The last stage reads the alphanumeric character and words by correlation template matching, which is a simple, fast and given a recognition rate reach to 98.245%.
