Scribd
Upload
7 views20 pages

Smart Parking System

The document outlines the implementation of a Smart Parking System using YOLOv8 and Automatic Number Plate Recognition (ANPR) technology. It details the methodology, data collection, and validation processes, achieving high accuracy in vehicle license plate detection. The system aims to improve operational efficiency, user experience, and traffic management in smart cities.

Uploaded by

Tanya Maheshwari
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
7 views20 pages

Smart Parking System

The document outlines the implementation of a Smart Parking System using YOLOv8 and Automatic Number Plate Recognition (ANPR) technology. It details the methodology, data collection, and validation processes, achieving high accuracy in vehicle license plate detection. The system aims to improve operational efficiency, user experience, and traffic management in smart cities.

Uploaded by

Tanya Maheshwari
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/ 20

SMART PARKING

SYSTEM USING
YOLOV8:
A SMART CITY
TECHNOLOGY
CONTE
1.Automatic Number Plate Recognition
2.Working of ANPR

NTS
3.Problem Statement
4.Data Collection and Validation
5.Literature Review
6.Methodology
7.Implementation
8.Performance Scores of Yolov8
9.Results
10.What’s Next for Smart Parking?
11.Other Applications
AUTOMATIC NUMBER PLATE
RECOGNITION AZ123

Automatic Number Plate Recognition Used for reading the number plate of the High speed image capturing.
(ANPR), is a very secure and accurate vehicles without involving human
system. interaction.
WORKING OF ANPR

First camera takes an image of the Algorithm detects the license plate in Data of the vehicle is saved using
vehicle. image & OCR reads the letters and license plate as a key.
characters.
PROBLEM
STATEMENT 8

To detect vehicle license plate using To extract text from detected license Maintaining the records of Vehicle
YOLOv8. plate using pytesseract. License plate with Entry and Exit time
records.
8

DATA COLLECTION AND


VALIDATION
- Quantitative validation: Dataset of 433 images, tested for
accuracy.
- Qualitative insights: Stakeholder interviews, focus groups, and
user surveys. 8
- Results: High accuracy, robust scalability, and user-centric
design improvements.
Link to dataset :-
https://www.kaggle.com/andrewmvd/car-plate-detection
LITERATU
RE
LITERATURE
REVIEW
LITERATURE
REVIEW
METHODO
LOGY
YOLOv
8 METHODOLOGY
Smart Parking System Using YOLOv8
•Yolov8 Algorithm is trained with images
and license plate annotations.
•Predict the licence plate annotation for a
test image using the trained Algorithm.
•Apply segmentation(cropping) on the
detected licence plate.

• Apply OCR(pytesseract) over the


cropped license plate.
• Store the recognised characters in the
database with Entry time.
• Repeating the same process from 2-4 at
the time of exit of the vehicle.
• Stores the exit date and time and
calculate the respective amount.
IMPLEMENTA
TION
IMPLEMENTATI
Implementing Yolov8 Algorithm Save Licence plate data

ON
•Training over images and annotations. 1.Save the extracted licence plate data extracted from
•Predicting the annotations on test data. OCR.
•Algorithm predicts 5 components: 2.Store Entry date and time.
1.(x,y) represents the centre of the bounding 3.Store Exit date and time and calculate the cost
box. propotional to the difference in Entry and Exit time
2.(w,h) are the width and the height of the
boxes.
3.Confidence represents the Estimated
Prediction Accuracy of the object.

Segmentation
Cropping out the licence plate from the
vehicle image using the predicted
annotations.
Apply OCR using pytesseract
1.OCR recognises the characters present in
an image. CROPPED LICENCE
2.Apply OCR using pytesseract API available PLATE
in python modules.
DETECTION OF A LICENCE
EVALUATION SCORES OF THE
YOLOV8
Scores on Traing dataset.
•Score apprentice depicts the training
score.
•Score validation depicts validation
score.(works as a part of test data from
the same training dataset)
•The accuracy score on training data is
94.2% and accuracy score on validation
data is 80%.
RESULTS
DATABASE AT
ENTRY

DATABASE
AFTER EXIT
WHAT’S NEXT
FOR SMART
PARKING?
• Proper camera angle setup can lead to more
efficient and effective detection.
• The recognisation of charaters of licence plate
can be improved if the captured image of the
vehicle is highly defined.
• The project can be further extented by
implenting object detection of types of vehicle
i.e. detection of cars or bikes and continue with
the same procedure.
OTHER APPLICATIONS
IMPACT ANALYSIS
Operational Improvements
1.Real-Time Monitoring
• IoT cameras offer live tracking of parking spaces, optimizing
utilization.
2.Error Reduction
• Advanced AI (YOLOv8) delivers precise vehicle identification,
boosting system reliability.
3.User Experience
• Mobile apps streamline parking with slot reservations and
real-time notifications.
4.Traffic Management
• Reduced congestion through guided navigation to available
spots.
5.Sustainability
• Lower emissions contribute to eco-friendly urban
environments, supporting smart city goals.
THANK
YOU
SUBMITTED BY:

SAISH KOHLI (08713702022)

You might also like