GHOSTPLATE

Spectral Vehicle Recognition

MEMBERS
01. Roshan Dubey
02. Shadab Shaikh
03. Ayush Chaudhary
02
Introduction
·This project focuses on developing a Vehicle
Number Plate Recognition System that can
automatically detect and capture vehicle license
plate information using image processing
technologies. The system is primarily aimed at
enhancing security in gated communities, societies,
and roadways by tracking vehicle movement. In
addition to recording the license plate, the system
will capture the driver’s photo, providing an
additional layer of security. The project will
combine advanced Optical Character Recognition
(OCR) with a database system to store and manage
vehicle and driver information.
motivation
 Problem Statement
In many residential societies, parking lots, and controlled roadways, vehicle monitoring
and security are often managed through manual entry logs, CCTV footage, or RFID-
based systems. These methods, while effective to some extent, are prone to several
limitations:
1. Manual Errors: Human involvement in recording vehicle entry and exit can lead to
errors, making the process unreliable and time-consuming.
2. Security Gaps: Lack of real-time monitoring and automatic logging makes it difficult
to keep track of unauthorized vehicle entries or unusual activity, posing a security
risk.
3. Inefficient Record-Keeping: Current systems often do not integrate driver
identification or photograph storage, meaning any later verification of the driver can
be challenging.
4. Inability to Scale: As the number of vehicles increases, manually keeping track of
vehicles and drivers becomes inefficient, especially in large societies or busy
roadways.
5. Data Access: The traditional methods lack centralized, easily accessible, and
searchable records of vehicle movement, making it difficult to retrieve historical
next slidedata
for security audits or incidents.
 Objectives
1. Automated Vehicle Number Plate Recognition:
Develop a system that automatically captures the number plate of vehicles entering
or exiting a designated area using image processing techniques.
Use OCR (Optical Character Recognition) to extract the vehicle number from images.
2. Driver Identification and Security:
Capture and store the image of the driver for security purposes. This will provide a
record that can be cross-referenced in the event of any security incidents.
3. Entry and Exit Time Logging:
Automatically record the exact entry and exit time of each vehicle. This data will be
stored in a database and accessible for reporting and audits.
4. Database Management:
Develop a robust database system to store and manage information about the
vehicles (number plates), driver photos, and timestamps. The database should be
searchable, allowing for easy retrieval of historical data.
5. Real-Time Processing and Alerts:
Ensure real-time processing of the vehicle recognition system, providing instant data
next slide
logging. Alerts should be generated for unauthorized vehicles or if any pre-defined
criteria are violated.
 Architecture/Framework
1. Image Capture Module:
Cameras installed at entry and exit points capture images of vehicles and drivers.
2. Image Preprocessing:
Preprocessing techniques using OpenCV include grayscale conversion, edge detection, and noise
reduction to enhance the image quality for recognition.
3. License Plate Detection:
The number plate detection subsystem uses contour detection and segmentation techniques to
isolate the number plate from the vehicle image.
4. Optical Character Recognition (OCR):
Tesseract OCR is used to extract text (vehicle number) from the preprocessed images of number
plates.
5. Driver Identification:
A separate image capture system is used to take photos of the driver during vehicle entry and exit,
adding a layer of security.
6. Database Management:
A MySQL/SQLite database is employed to store vehicle numbers, driver images, and timestamps
(entry and exit times). This database can be queried for real-time monitoring, alerts, and historical
analysis.
7. User Interface:
A simple web-based or desktop interface for security personnel to view real-time data, search for
specific vehicles or drivers, and generate reports.
 Details of Hardware & Software
Hardware Components
1. Cameras:
High-resolution IP cameras (or similar) installed at vehicle entry and exit points.
2. Processing Unit:
A desktop or server-grade system with sufficient GPU and processing power to handle image
processing tasks in real-time.
3. Storage:
Local or cloud-based storage for storing captured images, videos, and the database records.
Software Components
1. Programming Language:
Python (or Java) for implementing the image processing algorithms and OCR integration.
2. Image Processing Library:
OpenCV for preprocessing and detecting vehicle number plates.
3. Optical Character Recognition:
Tesseract OCR for number plate recognition.
4. Database:
MySQL/SQLite for storing and managing vehicle information and images.
5. Web Interface:
HTML/CSS/JavaScript or frameworks like Flask/Django for building the user interface for monitoring
and administration.
 Conclusion and
Future Work
The proposed Vehicle Number Plate Recognition System successfully
addresses the limitations of existing systems by providing real-time
monitoring, driver identification, and efficient data storage. The
system is scalable and adaptable to various environments such as
societies, parking lots, and roadways.