Automatic License Plate Recognition System

Automatic License Plate Recognition System

Nima Sanga Tenzin Tshering Deker Tshewang Dechen Yudhistra Hang Subba

INDEX

Problem Statement

Background Information

Technology

Aim & Objectives

Literature Review

System Design

Scope of the Project

Requirements

Workflow

01

Problem statement

• In 2021, there were around 116,926 vehicles registered with the Road Safety and Transport Authority
• The increase in the number of vehicles, therefore, has also resulted in many problems, such as traffic congestion , shortage of parking spaces, and pollution
• The manual system of monitoring and regulating traffic involves tedious procedures which proves to be inefficient and time consuming
• To overcome this issue, we will use the Automatic License Plate Recognition System. This system will identify vehicle number plates using image processing algorithms and will easily automate the detection of vehicle number plates.

02

Aim & Objectives

Aim

• To develop a system which recognizes license number plates

Objectives

• To design an efficient and automatic license number plate recognition system.
• To automate the traffic management process.

03

Scope of the Project

• The scope of the project is within the country
• In this project, a model will be trained using computer vision, which includes Optical Character Recognition and Convolutional Neural Network, to recognize license number plates

• The model will then be deployed in a website wherein users can upload images to identify number plates.

04

Background Information

• Traffic management involves a wide range of activities such as traffic monitoring, traffic signal timing, monitoring traffic violations and accident response. These activities involve manual effort and consume a longer period of time
• License number plays a significant role in tracking and keeping records in all traffic activities
• Traffic management in our country is presently dependent on human labor as a result of which there can be potential errors while monitoring traffic related activities
• In areas where there are a larger number of vehicles, the traffic management has limited capacity to monitor traffic flows and vehicle movements

04

Background Information

• We proposed this project with the aim to automate and streamline processes involved in traffic management by developing a system which detects vehicle number plates
• This system can be applicable for monitoring traffic as well as for security purposes

05

Literature Review

Vehicle Registration Plate Recognition System Using Template Matching

• According to this research, Registration plate recognition is widely used in detecting speedy cars, traffic law enforcement and electronic toll collection
• The problems associated with registration plate recognition are, plate images have different quality, illumination, view angle, distance, complex background and fonts
• To address these problems, image processing tools are used.

05

Literature Review

Automatic License Plate Detection and Recognition Using Deep Learning

• In this paper, it discusses an autonomous information system to reform vehicle information systems in real time.
• The system of vehicle number plate detection and recognition detects the number plate and extracts texts from the image using location algorithms, plate segmentation and character recognition
• The system is applicable in command force, parking management,and road safety
• According to this paper, they concluded that the performance can be improved by increasing the number of hidden layers of neurons when using the Multi-layer Perceptron Classifier and when the nearest neighbor number is increased while using the KNN.

06

Requirements

Functional Requirements

• Upload Image
• Detect & recognize license number plate
• Display Result

Non - Functional Requirements

• Usability
• Portibility

07

Technology

• The software technologies that will be used for the system are as follows
• Visual Studio Code
• React JS
• GitLab
• TensorFlow

The hardware technology that will be used for the system are as follows

• Laptop with 8GB RAM minimum, 16GB RAM recommended

07

System Design

Application layer

Presentation layer

Data layer

• The user uploads an image of any vehicle number plate
• This is followed by the process of image recognition and image processing.The extracted vehicle number will then be presented to the user.

• The uploaded image will go through processes such as image recognition, image processing, character recognition and post-processing where the recognized image will then be verified and validated against the known format.

• The data tier or database tier will store the extracted vehicle number

08

Workflow

The End