[Paper Review] Proposal for Automatic License and Number Plate Recognition System for Vehicle Identification

Hamed Saghaei|arXiv (Cornell University)|Oct 9, 2016

Vehicle License Plate Recognition6 references16 citations

TL;DR

This paper proposes a computer vision-based automatic license and number plate recognition (LNPR) system that uses image processing to detect and recognize vehicle license plates from camera-captured images without requiring additional hardware like GPS or RFID. The system employs localization, normalization, segmentation, and OCR techniques, achieving successful plate detection and recognition on real-world images, with demonstrated applicability in security and traffic control applications.

ABSTRACT

In this paper, we propose an automatic and mechanized license and number plate recognition (LNPR) system which can extract the license plate number of the vehicles passing through a given location using image processing algorithms. No additional devices such as GPS or radio frequency identification (RFID) need to be installed for implementing the proposed system. Using special cameras, the system takes pictures from each passing vehicle and forwards the image to the computer for being processed by the LPR software. Plate recognition software uses different algorithms such as localization, orientation, normalization, segmentation and finally optical character recognition (OCR). The resulting data is applied to compare with the records on a database. Experimental results reveal that the presented system successfully detects and recognizes the vehicle number plate on real images. This system can also be used for security and traffic control.

Motivation & Objective

• To develop a low-cost, automated system for identifying vehicles using license plate recognition without relying on GPS or RFID.
• To enable real-time vehicle identification in traffic monitoring and security applications through image processing.
• To address challenges in license plate detection under varying lighting, angles, and image quality conditions.
• To design a modular pipeline integrating localization, normalization, segmentation, and OCR for robust recognition.
• To validate the system's effectiveness using real-world vehicle images and compare results against database records.

Proposed method

• Utilizes specialized cameras to capture images of passing vehicles for input to the LNPR system.
• Applies image preprocessing techniques to enhance contrast and reduce noise in captured images.
• Employs a localization algorithm to detect the region of interest (ROI) containing the license plate in the image.
• Uses orientation correction to normalize the plate's angle for consistent processing.
• Applies normalization to standardize plate size and lighting conditions across different images.
• Performs segmentation to isolate individual characters on the plate before feeding them to the OCR engine.

Experimental results

Research questions

• RQ1 Can a vision-based system reliably detect license plates in real-world traffic images without additional sensors?
• RQ2 How effective is the proposed pipeline in handling variations in lighting, angle, and image quality?
• RQ3 To what extent does the system achieve accurate character recognition using standard OCR techniques?
• RQ4 Can the system be deployed for real-time traffic monitoring and security applications?
• RQ5 What is the recognition accuracy of the system when tested on diverse, real-world vehicle images?

Key findings

• The proposed LNPR system successfully detects and recognizes license plates in real-world images under various conditions.
• The system achieves high accuracy in plate localization and character recognition using a combination of image processing and OCR techniques.
• No additional hardware such as GPS or RFID is required, reducing implementation cost and complexity.
• The system demonstrates robustness in handling variations in lighting, orientation, and image resolution.
• Experimental results confirm the feasibility of using image-based processing for automated vehicle identification in traffic and security contexts.
• The system's modular design allows for integration into existing surveillance and traffic management infrastructures.