Smart Parking System using Image Detection Algorithms & Optical Character Recognition (OCR)

Abstract

The Smart Parking System is an innovative and effective solution designed to make parking more efficient in urban areas, using advanced technology to help both vehicle owners and parking operators automate and optimize the parking process. When a vehicle enters the parking lot, it is successfully identified using an advanced sensor and camera system, with a high-resolution camera that records the license plate number and optical character recognition (OCR) technology. Get accurate license plate numbers. The data is then stored in the MongoDB database along with the input timestamp and when the vehicle moves, the license plate is re-identified using the same technique and the system stores the input timestamp associated with the detected license plate in the database. he does. Reading from Parking time is calculated from the time difference between entering and leaving the parking lot, and the parking fee is calculated per hour. To facilitate payment, a QR code is generated and displayed at the exit. Vehicle owners can use their mobile device to scan this QR code and connect to the secure payment gateway. After the payment is completed, the exit door will open automatically and the car can leave the parking lot. This smart parking system offers many benefits, including fewer numbers, increased accuracy, reduced waiting time, and efficient payment processing. It reduces congestion in urban areas and improves traffic flow while providing users with convenient and trouble-free parking.

Introduction

I. INTRODUCTION

Ever-increasing urbanization and exponential growth in the global population have led to an urgent need for smart and efficient urban infrastructure solutions. Among the myriad challenges posed by rapid urban development, one of the most widespread and vexing problems is urban mobility and the lack of parking spaces in densely populated areas. Among the effects of inadequate parking facilities, we can mention traffic congestion, increase in fuel consumption, increase in environmental pollution, and most importantly, passenger dissatisfaction.[1][2] Parking management and optimization of parking resources have received considerable attention in recent years. One of the most promising and technologically advanced solutions to this problem is the implementation of intelligent parking systems (SPS). This research paper explores the development and deployment of an SPS that uses cutting-edge technologies such as advanced sensors, high-resolution cameras, optical character recognition (OCR) and secure payment gateways to create a seamlessly integrated and efficient parking experience. for city dwellers.

In the field of computer vision and artificial intelligence, image detection is a pivotal and dynamic field. At its core, image detection is the process of identifying and locating objects or features of interest within a digital image or video. This technology allows machines to interpret and understand the visual world much like the human visual system, but with the unparalleled ability to process vast amounts of visual data at incredible speeds. The importance of image detection lies in its wide applications in various fields, from healthcare and autonomous vehicles to security and retail. It has changed the way we interact with technology and opened the door to a new era of innovation and automation.

Optical character recognition (OCR) is an innovative technology that bridges the gap between the analog and digital worlds, allowing printed or handwritten text to be converted into machine-readable text.[5] PyTesseract, a Python library, brings OCR capabilities to developers and provides a versatile and easy-to-use interface for extracting text from images, scanned documents, and even live cameras. For our implementation, we used PyTesseract, an optical character recognition (OCR) engine developed by Google. It is based on the Tesseract OCR engine, originally developed by Hewlett-Packard Labs and later maintained by Google. The PyTesseract library provides a Python interface to the Tesseract OCR engine, allowing developers to easily extract text from scanned images and documents using Python.

In the following sections of this paper, we delve into the complexities of the smart parking system, exploring the underlying technologies, system architecture, and its transformative impact on urban spaces. Furthermore, we will assess the practical implementation of this system and discuss possible future improvements and scalability. As we continue, the ingenuity of the Smart Parking System will become increasingly apparent, reaffirming its potential to reshape the urban landscape in pursuit of greater efficiency and convenience.

II. BACKGROUND

A. System Overview

Smart Parking System (SPS) is an innovative solution designed to automate and optimize urban parking. It uses advanced technology, including high-resolution cameras and OCR software, to detect vehicles, capture license plate data and calculate parking fees. SPS aims to create an efficient self-service system that reduces traffic congestion, improves urban mobility and improves the overall parking experience.

B. Internal Interface

The core of the system, the backend, is responsible for the seamless coordination and management of various components and ensures the smooth flow of data and processes. Backend processes include data processing, storage and retrieval, all synchronized in real-time to maintain accurate records of vehicle entry and exit. In addition, it plays a key role in the secure generation of QR codes for payment processing and offers a reliable tamper-proof gateway. These complex backend operations are necessary for the Smart Parking System to provide a reliable and efficient parking solution that provides a seamless experience for city dwellers.

C. Client Side Interface

Our client-side smart parking system control panel provides an intuitive interface for vehicle owners and operators. It offers quick access to essential information, including their number plate, entry time and a visual representation of parking fees. Users can easily track parking time and related charges through a user-friendly spreadsheet that ensures full transparency and easy payments. This dashboard provides users with real-time insight and facilitates hassle-free parking, making it a valuable tool for improving urban mobility and efficiency.

III. RELATED WORK

Several research papers have played an important role in shaping the development of smart parking systems. The research "Efficient number plate recognition and recognition" [6] by Anurag Singh Rawat, Himanshu Durrani, Aman Yadavvanshi, Manvi Bohra, Inderjit Kumar and Tekam Singh contributed significantly to the accuracy and efficiency of the system. In this study, we present an automatic number plate recognition (ANPR) system that effectively detects and recognizes license plates and achieves an impressive accuracy of 97.0%.

This paper shows the potential to improve ANPR systems by integrating object recognition and optical character recognition (OCR) technology, paving the way for more efficient and accurate parking.

Additionally, the research "Vehicle Sign Recognition for Law Enforcement" [1] by Ananya Sri Shetty, V Sai Vineetal, Sreya Ravi, Nerella Likhitha, and Dr. K Anuradha has helped facilitate the operation of smart parking systems. This study highlights the importance of Automatic Vehicle Number Recognition (AVPR) in law enforcement and crime prevention in urban areas. We present an AVPR model that uses live streaming to evaluate parameters such as image size, success, and processing time to solve problems such as parking traffic congestion and improve vehicle safety.

In addition, the document "Urban Traffic Control with Vehicle Identification Plates" [3] by Professor K.P. Kamble, Prof. Rachana Admane, Prof. Ritika Shekar, Prof. Kalyani Chaudhary, and Prof. Samiksha Wagde provided valuable insights into the development of smart parking systems.

This study focuses on Vehicle Number Plate Recognition (VNPR) and investigates the localization and character recognition steps of license plate processing. This study's approach to character segmentation and recognition using optical character recognition shows the potential of the system for applications such as toll fields. Together, these research findings form the architecture of an intelligent parking system, making it a valuable solution for efficient and automated urban parking.

IV. PROPOSED SYSTEM

A. System Overview

The main goal of the proposed smart parking system is to automate this process and make it hassle-free. The proposed system aims to revolutionize the traditional paradigm of parking management by introducing an OCR-based intelligent parking system. This innovative solution uses optical character recognition (OCR) technology to automate the parking entry and exit process.

At the entry gate, the system uses a camera and OCR software to capture and identify the license plate and stores this information in the MongoDB database along with the entry time stamp [4][7]. This data serves as the basis for the entire parking operation. Exit gates similarly use OCR technology to read license plates, which facilitates real-time matching with incoming records and enables accurate billing based on hourly rates.

By generating QR codes for seamless payments, the system enhances the user experience and eliminates the need for manual intervention. The main goal of this project is to create an unmanned, efficient and cost-effective smart parking system that will benefit both parking operators and users.

This system not only promises to simplify parking operations, but also offers a glimpse into the future of automated, convenient and seamless parking management in urban environments. By eliminating manual intervention, it provides a cost-effective, efficient and easy-to-use solution for parking operators and drivers.

B. System Architecture

1. The system architecture diagram shows three primary entities: camera, database, payment.
2. The entity camera detects the vehicle and the license plate.
3. Using OCR, we extract license plate characters and store them in a database with a timestamp.
4. We use MongoDB to store license plates and timestamps.
5. On output, we compare input and output timestamps and generate bills on an hourly basis.
6. To facilitate payments by providing a QR code for payment integration.

C. Raspberry PI Integration

To enhance the demonstration of our OCR-based intelligent parking system, we integrated a Raspberry Pi, a versatile and compact single-board computer, into the system architecture. After successfully detecting and storing the vehicle license plate and input data in MongoDB, the Raspberry Pi comes into play. One practical demonstration of the system's capabilities involves using a Raspberry Pi to control LEDs. After the vehicle has successfully entered and recorded its data, the Raspberry Pi receives a signal from the system and in response activates an LED that visibly signals the entry event.

Similarly, after completion and successful payment, another LED connected to the Raspberry Pi blinks to indicate the completion of the parking process. This intuitive and visual feedback system not only engages users, but also adds another layer of transparency and satisfaction to parking. By incorporating the Raspberry Pi into our demonstration, we highlight the system's real-time capabilities and demonstrate the seamless, automated and unattended nature of our smart parking solution.

D. Use of MongoDB for Storage

Smart Parking System relies on MongoDB as its central database due to the need for a scalable and flexible storage solution. MongoDB's NoSQL architecture enables efficient storage and retrieval of large volumes of data. Its horizontal scaling capabilities ensure that the system can easily adapt to increasing data loads. MongoDB's flexibility also allows for dynamic data structures that allow storing different types of data associated with vehicles, their license plates, and payment records.

The database provides robust authentication mechanisms allowing access control down to the document level. This means that only authorized personnel can edit or access specific data, ensuring that sensitive information such as number plate details and payment records remain protected. MongoDB's authentication features, including role-based access control and integration with various authentication protocols, offer a comprehensive security framework to protect data from unauthorized access and potential breaches. As a result, vehicle owners and operators can trust that their personal and payment information is securely stored and processed in the system.

MongoDB is vital for instant storage and retrieval of vehicle entry and exit records Its support for indexing and aggregation ensures fast data processing and enables quick responses to user requests. This real-time processing capability is integral to creating smooth and hassle-free parking for users. The document-oriented nature of the database lends itself well to unstructured and constantly changing data related to parking facilities.

To ensure continuity of services, MongoDB provides built-in redundancy and high availability features. Databases can be configured to replicate data across multiple servers and locations, providing fault tolerance and disaster recovery capabilities. This means that even in the event of hardware failure or network issues, our smart parking system remains operational, minimizing disruptions for users and parking operators alike.

E. Innovation

One of the main innovations that define this OCR-based intelligent parking system is its unwavering commitment to achieving an unattended parking management solution. Traditional parking facilities have long relied on manual intervention, from ticketing to payment processing. This system is revolutionizing the industry by automating the entire process, from entry gate to exit, without the need for a single human. This innovation not only drastically reduces operating costs for equipment owners, but also offers an excellent and hassle-free experience for drivers. Unmanned access opens up possibilities for 24/7 operation where parking is available at any time, all while minimizing labor costs and increasing efficiency.

Another remarkable innovation in our system is the integration of seamless payments and increased user convenience. With OCR technology and real-time data management, the system calculates parking charges based on hourly rates and generates QR codes for payment. This eliminates the need for physical payment kiosks or manual transactions, providing drivers with an easy and contactless payment experience. In addition, the system enhances user convenience through visual indicators such as flashing LEDs to confirm entry and exit, making the parking process more transparent and user-friendly. This innovation not only saves time and reduces driver friction, but also ranks the system as a promising solution for modern urban environments.

Beyond the obvious automation and user-centric features, the OCR-based smart parking system harnesses the power of data-driven decision-making. The system collects a large amount of data, including entry and exit timestamps, vehicle license plates and toll calculations. This data is not only used for accurate billing, but also offers valuable information on parking facility usage and trends. The system allows operators to track occupancy, peak times and customer behavior, allowing them to optimize traffic and adjust pricing strategies as needed. Data-driven decision making is a powerful innovation that transforms parking facilities into more adaptable, efficient and profitable assets that align with the evolving needs of the urban environment.

Conclusion

In conclusion, this research paper presents a comprehensive this innovative system, designed with the primary goal of achieving unattended operation, offers a myriad of benefits that not only meet the demands of the modern urban environment, but also anticipate and prepare for the future of Parking Management. The unattended functionality of this system represents a fundamental shift in the way parking facilities are managed. By seamlessly automating input and output processes, the system reduces dependence on human service and consequently reduces labor costs while simultaneously increasing operational efficiency. Eliminating manual interventions results in more cost-effective and user-oriented parking. Drivers benefit from reduced waiting times, seamless payments and transparent entry and exit confirmations through visual indicators. In addition, the system supports round-the-clock traffic, making parking available at any time, adapting to the dynamic nature of urban life. In addition to its unattended nature, the system implements seamless payment processes and uses OCR technology to generate QR codes for easy, contactless payments. The system\'s data-driven approach offers additional benefits, allowing operators to gain insight into occupancy, peak times and customer behavior. This information allows operators to optimize operations, adjust pricing strategies and transform parking facilities into adaptive, efficient and generating assets. As urban areas continue to grow and evolve, this innovative solution is at the forefront of parking management technology, laying the foundations for a more efficient, convenient and cost-effective future. Overall, this research paper contributes to the development of intelligent parking systems, offering unattended design, seamless payment processes and data-driven decision-making, making it a valuable addition to the arsenal of tools available to urban planners, parking facilities. operators and drivers. The project represents a successful step towards solving the challenges of modern urban life and at the same time provides a vision for a smarter, more efficient and user-oriented urban infrastructure.

References

[1] Ananya Sri Shetty,V Sai Vineetal,Sreya Ravi, Nerella Likhitha, Dr K Anuradha. \"Vehicle Number Plate Detection through live stream using OCR\" (2023). [2] Sarthak Ranjan,, Shrusti Sanket, Shivam Singh, Shivam Tyagi, Mandeep Kaur, Nitin Rakesh, Parma Nand. (2022).OCR based Automated Number Plate Text Detection & Extraction [3] Prof.K.P.Kamble, Rachana Admane. Ritika Shekar. Kalyani Chaudhary,Samiksha Wagde. \"Image Processing based Vehicle Number Plate Recognition,\" in IEEE international conference on I-SMAC (2022) [4] Sparsh Jain, Rishikesh Rathi, Rahul Kumar Chaurasiya. (2018). Indian Vehicle Number Plate Recognition using Single Shot Detection & OCR.IEEE India Council International Subsections Conference (INDISCON)(2021) [5] Sarthak Ranjan,Shrusti Sanket,Shivam Singh,Shivam Tyagi,Mandeep Kaur,Nitin Rakesh,Parma Nand. “OCR based Automated Number Plate Text Detection and Extraction” 2022 9th International Conference on Computing for Sustainable Global Development (INDIACom) [6] Anurag Singh Rawat,Himanshu Devrani,Aman Yaduvanshi,Manvi Bohra,Indrajeet Kumar,Teekam Singh.”Surveillance System using Moving Vehicle Number Plate Recognition”. 2023 2nd International Conference on Edge Computing and Applications (ICECAA) | 979-8-3503-4757-9/23/$31.00 ©2023 IEEE | DOI: 10.1109/ICECAA58104.2023.10212426 [7] Wichan Thumthong, Phayung Meesud, Pita Jarupunphol. “ Automatic Detection & Recognition of Thai Vehicle License Plate from CCTV Images”.2021 13th International Conference on Information Technology and Electrical Engineering (ICITEE) | 978-1-6654-4306-7/21/$31.00 ©2021 IEEE | DOI: 10.1109/ICITEE53064.2021.9611826 [8] Hassanin M. Al-Barhamtoshy,Muhammed K. Dauda,Khalid K. Abdullah,Tarik F. Himdi. “Detecting Available Parking Spaces in Smart Cities”. 2023 1st International Conference on Advanced Innovations in Smart Cities (ICAISC) | 978-1-6654-7275-3/23/$31.00 ©2023IEEEDOI:10.1109/ICAISC56366.2023.1008538

Copyright

Copyright © 2023 A. Paranjape , N. Mahangade, A. Nikam, B. Rajekadam, Mr. J. R. Chavan . This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.