Scribd
Upload
36 views10 pages

Speed Based Automatic Vehicle's Number Plate Recognition System

This document summarizes a student project on developing a speed based automatic vehicle number plate recognition system using YOLOv5. The system aims to automatically detect traffic violations like vehicles crossing signals during red lights through cameras installed at traffic junctions. This would help enforce traffic rules more efficiently as compared to the existing manual system. The proposed system uses deep learning algorithms for image processing, license plate detection and character recognition to identify rule violators and issue fines electronically.

Uploaded by

SRIRAM P
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
36 views10 pages

Speed Based Automatic Vehicle's Number Plate Recognition System

This document summarizes a student project on developing a speed based automatic vehicle number plate recognition system using YOLOv5. The system aims to automatically detect traffic violations like vehicles crossing signals during red lights through cameras installed at traffic junctions. This would help enforce traffic rules more efficiently as compared to the existing manual system. The proposed system uses deep learning algorithms for image processing, license plate detection and character recognition to identify rule violators and issue fines electronically.

Uploaded by

SRIRAM P
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 10

SPEED BASED AUTOMATIC

VEHICLE’S NUMBER PLATE


RECOGNITION SYSTEM
USING YOLOV5
BY

P. KAVISH GOLDEN(1912071)
P. SRIRAM(1912405)

Guided by:
Ms.Priyadharshini, AP,CSE
INTRODUCTION:

Traffic rules are important for safety as traffic laws are to prevent drivers of vehicles from causing
accidents or hitting pedestrians. Traffic Rule Violations are leading cause of accidents, according to WHO India
is a leading country in casualties occurring on road. The current system uses human interaction for rule violation
detection, as it is a manual process it has some limitations. An alternative solution would be AI-developed
System. With our system, we can detect multiple rule violations, for example, Vehicle crossing signal during red
light etc. Basic idea is to detect these violations through preinstalled cameras. We can do it by ML based
algorithm where we can detect the violators by Image Processing, getting the number plate, categorizing
violation accordingly and issuing fine. Which will help increase the efficiency of traffic rule enforcement.
ADVANTAGES:

• In earlier days there was only cameras available to detect the criminals.
• People have to watch the monitor continuously to reduce crime.
• But now using this system it is easy to detect crimes.
• If this system gets implemented, we can reduce the number of the below cases drastically,
1.Hit and run
2.Drunken driving
3.We can bring justice to the victims and can even solve the pending cases provided the videos
are available.
EXISTING SYSTEM

Currently in many sectors vehicle is identified using the number plate which is manually noted
by a human which is a slow process. Thus, vehicle number plate recognition is an intensive manual
process which can perhaps be automated using deep learning which forms the basis of this project.
This is all about the existing system of the project“ Speed based automatic vehicle’s number plate
recognition system Using yolov5”
EXISTING SYSTEM DRAWBACKS

• More Manpower.

• More Expenditure.

• High Error Probability.


• Less Performance.
• More Dependency.
• Time Taking.
SOCIAL NEED:

• Vehicle speed detection was a crucial parameter for the system as the city CCTV cameras are installed at
different heights and angles. So, our team of developers did some brainstorming and came up with a
customized geometry mathematical solution that can calculate the speed of the vehicle considering all the
parameters of the camera.
• The existing CCTV cameras are used to cover the entire range of the city road including the side road
traffic that used to create glitches in the system. So, we designed a feature that allows the admin to draw the
boundaries of the road map. The cameras will only monitor traffic inside those boundaries of the road map.
• The existing CCTV cameras are used to cover the entire range of the city road including the side road
traffic that used to create glitches in the system. So, we designed a feature that allows the admin to draw the
boundaries of the road map. The cameras will only monitor traffic inside those boundaries of the road map.
DISADVANTAGE

• More Manpower.

• More Expenditure.

• High Error Probability.


• Less Performance.
• More Dependency.
• Time Taking.
IMPLEMENTATION:

• Loading and data pre-processing


• Training the model
• Estimation of performance
• Image Desaturation
• Image Thresholding
• Gaussian Filter
• Morphological Transformation
• Segmentation
• Character Recognition
OUTCOME

• Python model training to detect and recognize license number plate.

• The below output is achieved by running the python program which trains the model using the

• training dataset and saves the model by using python libraries. Then a predictor file is run with a

• test image which is not included in the training dataset and gets recognized and for better

• visibility of detection the license plate area is marked by a red border.

• This program can be used for variety of works where a license number is the primary way of

• recognizing a vehicle. Current application is completely usable for little uses as parking area or

• traffic line use.

• In future we can include face acknowledgment and connection the entire framework with

• criminal database and recognize passing criminal outlaws

You might also like