Automatic Number

Plate Recognition
INDIAN INSTITUTE OF TECHNOLOGY
PATNA BIHTA - 801106,INDIA
Student name-Raju Ranjan

Roll No. 12res24a1041

Group No.70
CANDIDATE’S DECLARATION

Date-28/04/2025

I hereby declare that this submission is my own work and

that, to the best of our knowledge and belief, it contains no
material previously published or written by another person
nor material which to a substantial extent has been
accepted for the award of any other degree or diploma of
the university or other institute of higher learning, except
where due acknowledgment has been made in the text.

Date: 28/04/2025

Student Name-Raju Ranjan

Roll No- 12res24a1041

Group No.70

Signature of student
 My team Overview
Team Members
1. Aashish Ranjan : my teammate developed a visualization
module using matplotlib to display detected number plates
with bounding boxes. This helps in verifying the OCR output
and improves the clarity of results.
2. Raju Ranjan(MY SELF): My name is Raju Ranjan. I am
working on the image processing part of the project using
OpenCV. I focuses on detecting and extracting number plate
regions from vehicle images and also assists in OCR-related
tasks when required.
3. Pranav Ranjan : My team mate on implementing Easy OCR
to accurately extract text from the detected number plates.
my main focus is on the OCR part but i also contribute to the
image processing tasks using OpenCV whenever needed to
ensure smooth integration between detection and
recognition.
4. Aaditya Ranjan : (No Response)
5. Piyush Ranjan : (No Response)
 Summary of project
In this project we are making Automatic Number Plate Recognition (ANPR)
system which can automatically detect and read vehicle number plates from
stored images or live videos . Using Python, OpenCV, and EasyOCR, the
system is designed to identify number plates and extract the text from them
and store it in the database. This technology is useful in many real-life
situations, like monitoring vehicles at gates, keeping track of entries and exits
in parking lots,to observe vehical to see they are following traffic rule and aslo
make e challan . The aim was to build a easy to use and implement System .
 Table of Contents
Chapter Description Page
No. no.
1
1 Introduction and Problem Statement
2
2 My work
3
3 Future Work Plan
4
4 Conclusion
References 5
Chapter 1
Introduction and problem
1.1 introduction:
Automatic number plate recognition (ANPR) is technology that use to detect automatically
and vechical number plate from image or video.This technology is delevop to identify the
vehicles.ANPR is used to traffic monitoring,toll collection and for parking management.

This project aims to design and implement a reliable ANPR system that can automatically
process vehicle images, accurately detect number plates, and recognize the characters
with high accuracy. By integrating tools like OpenCV,machine learning models, and OCR
technologies, the system showcases the practical application of artificial intelligence and
computer vision in solving real-world transportation and security challenges.

1.2 Problem Statement :

In today's fast-paced and security-conscious world, there is a growing need for automated
systems that can monitor and manage vehicle movement.

Manual identification and tracking of vehicle number plates are time-consuming,difficult,

and impractical, especially in high-traffic areas.

The problem lies in developing a system that can automatically detect, extract,and
recognize vehicle number plates from images or live video streams, under various
conditions such as poor lighting, motion blur. The system must be capable of accurately
identifying number plate characters in real time

Thus, the goal of this project is to design an Automatic Number Plate Recognition (ANPR)
system that automates the process of vehicle identification, improves operational
efficiency, reduces human intervention, and supports applications like traffic law
enforcement, automated toll collection,parking management, and security surveillence.
1
Chapter 2
My work
1.Image Loading : The input vehicle image was loaded using cv2.imread() function.

2.Pre-Processing : Before detecting the number plate, several pre-processing operations

were performed:

a.Grayscale Conversion : The input color image was converted into grayscale to
reduce computational complexity.

b.Noise Reduction using Gaussian Blur : A Gaussian blur filter was applied to
the grayscale image to Remove Noise.

3.Edge Detection : To highlight the edges and boundaries in the image, Canny Edge
Detection was applied.

4.Contour Detection : Contours were found using cv2.findContours() function. Contours

help identify objects based on their shapes and boundaries.

5.Plate Region Extraction : Once the correct contour (rectangle) was found, the number
plate region was cropped from the original image using OpenCV’s bounding rectangle
functions.
2
Chapter 3
Future Work Plan
1. Real-Time Video Processing and Optimization:

● Objective: Improve the system’s ability to handle video feeds in realtime,ensuring fast
and accurate license plate detection from continuous video frames.

● Tasks:

○ Optimize frame extraction and processing speeds to handle high frame rates frames per
with minimal lag.
○ Investigate hardware acceleration to enhance performance during real-time processing.

○ Implement a more efficient video processing pipeline that can dynamically adapt varying
video qualities (resolution, frame rate) and formats.

2. Multi-Lane and Multi-Vehicle Detection:

● Objective: Extend the system’s capability to handle multiple lanes and vehicles in a
single frame, such as in toll booths or parking lots.

● Tasks:

○ Implement algorithms to detect and track multiple vehicles simultaneously, ensuring

accurate number plate detection from each Vehicle.

○ Improve the system’s ability to handle overlapping vehicles or closely packed plates,
which can often cause detection errors.

○ Test the system’s scalability for environments with heavy traffic flow or high-density
parking areas.
3
Chapter 4
Conclusion
In this project, we successfully developed an Automatic Number Plate Recognition (ANPR)
system using OpenCV and EasyOCR in Python. Through a combination of image
processing techniques and optical character recognition, the system was able to:
• Detect the number plate region from vehicle images,
• Extract and segment the plate area,
• Recognize and interpret the alphanumeric characters on the plate.

4
 References
1. OpenCV Documentation
. OpenCV - Open Source Computer Vision Library. from
https://opencv.org/

2. YouTube Tutorial: ANPR with OpenCV and EasyOCR

Programming Knowledge. ANPR (Automatic Number Plate Recognition)
with OpenCV and EasyOCR in Python Video. YouTube.
https://youtu.be/NApYP_5wlKY?feature=shared