Smart Barrier Control System by Vehicle Number Plate Detection Using ML and IoT
With many applications in security, surveillance, and intelligent transportation systems, automatic license Plate Recognition is a fundamental and important task. In recent era with the growth of Deep Learning techniques, ALPR also has shown dramatically accurate results as other computer vision fie
2025-06-28 16:35:06 - Adil Khan
Smart Barrier Control System by Vehicle Number Plate Detection Using ML and IoT
Project Area of Specialization Electrical/Electronic EngineeringProject SummaryWith many applications in security, surveillance, and intelligent transportation systems, automatic license Plate Recognition is a fundamental and important task. In recent era with the growth of Deep Learning techniques, ALPR also has shown dramatically accurate results as other computer vision fields, focusing on country-specific plates like Chinese, Indian, Brazilian, American or European. However, most of the researchers either focus on one task of ALPR pipeline or they are commercial and private dataset is used ultimately lacking the detailed information for other researchers.
In our project which is Smart barrier control system by vehicle number plate detection using ML and IoT, we propose four major tasks Front/Back view detection, License Plate Detection Character Segmentation and Character Recognition on Pakistani License Plates using self-built dataset. We are approaching YOLO for to detect Front/Back view of car. Faster-RCNN had shown best accuracy in Front/Back view detection of Number in many literatures.
We will use Pixel Counting algorithm for Character Segmentation for both standard and non-standard number plates. The final step was correct recognition of characters; we will develop our own deep neural network. In the last we give ha prototype of project with hardware and with web app and mobile app.
Project ObjectivesIn this project we are presenting a smart barrier control system which can detect vehicles and their number plates. This system will automate the manual process in which no human supervision will be required to open and close the gate or barrier. This system will create a daily inventory of registered vehicles and record will be available for any further analysis. Furthermore, there will be no need to introduce him/herself to guard for opening the barrier so by using this system a lot of time will be saved.
Automatic Number Plate Recognition is a computer vision technology that efficiently identifies vehicle number plates from images without the need for human intervention. In recent years, it has become more and more important due to three main factors: the growing number of cars on the roads, the rapid development of image processing techniques and the great quantity of real-life applications that this technology offers. Some of the most typical applications of ANPR systems are traffic law enforcement, automatic toll collection or parking lot access control. But this technology is also widely used for other, perhaps, more inspiring purposes like crimes resolution, as it helps to identify the cars of the offenders.
Project Implementation Method
In our project, we will propose a system consisting on four steps Car detection, License No. Plate Detection, Character Segmentation, Character Recognition. First of all, the frame (an image) is taken from camera image and first step to detect car is applied after that the cropped image is fed to our second step in order to detect vehicle number plate. The third step deals with the segmentation of characters, individual characters are extracted from plate and in 4th step each character image is fed to our deep neural network and recognized. And after number plate recognition send to the server where number plate inventory and all process will perform on number plate. The flow chart of IoT part given bellow.
We are approaching YOLO for to detect Front/Back view of car. Faster-RCNN had shown best accuracy in Front/Back view detection of Number in many literatures. We will used Pixel Counting algorithm for Character Segmentation for both standard and non-standard number plates. The final step was correct recognition of characters, we will develop our own deep neural network.
On the other hand, database will be created for storing daily basis data and web app will be created to show a daily inventory of vehicles. Queries will be received from microprocessor and will be sent to database and then answer will be received by Raspberry Pi to operate barrier in real time. In our prototype we make a parking system in which we automate the system and ensure the security. Finally, at the entry station, the extracted characters i.e., license plate number and entry time are stored in the entry level number plate database.
At the exit station, number plate also extracted, the extracting license plate number and exit time are stored in the exit level number plate database. Using template matching algorithm, the characters of the entry and exit number plates are compared with the help of correlation, and make daily base inventory and we have also developed web app and android app thorough which we can access all data regarding vehicle entrance and exit timings anywhere and any time.
This real time system will be able to:
- Detect incoming vehicles at the entrance of factories or institutions.
- Detect and read the number plates of the entering vehicles of university by image processing and machine learning/AI.
- Open the barrier for already registered vehicles in database using wireless System.
- Generate the alarm for non-registered vehicles to security guard for manual checking and opening of the barrier.
- Create a daily inventory for incoming and outgoing vehicles of university.
1) Car is detected using YOLO with good efficiency.
2)Detecting number plate of car using YOLO tensorflow1.5 with the help of machine learning
3)Number plate recoginition is done by using Tesrect OCR Algorithm.
4)Through WebAPP which we have developed with the help of MS SQL SERVER and Visual Studio for storing and monitioring vehicle data .
Final Deliverable of the Project Hardware SystemCore Industry TransportationOther IndustriesCore Technology Artificial Intelligence(AI)Other TechnologiesSustainable Development Goals Industry, Innovation and InfrastructureRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 40100 | |||
| Arduino | Equipment | 1 | 650 | 650 |
| Node MCU | Equipment | 1 | 550 | 550 |
| Barrier | Equipment | 1 | 10000 | 10000 |
| DC MOTOR | Equipment | 1 | 3000 | 3000 |
| DC POWER SUPPLY | Equipment | 1 | 1700 | 1700 |
| DC BATTERY | Equipment | 1 | 3000 | 3000 |
| CAMERA | Equipment | 2 | 8000 | 16000 |
| SENSORS,WIRES | Equipment | 1 | 1000 | 1000 |
| WIFI ROUTER | Equipment | 1 | 3000 | 3000 |
| Remote(Receiver+ Transmitter) | Equipment | 1 | 1200 | 1200 |