Automated Vehicle Profiling System
Due to the advanced development in the last few decades, License Plate Recognition (LPR) systems are quite popular. LPR systems are being developed and implemented by many countries due to the increasing number of traffic violations and crimes, these systems helps in the Automated Ticketing, Automat
2025-06-28 16:30:24 - Adil Khan
Automated Vehicle Profiling System
Project Area of Specialization Artificial IntelligenceProject SummaryDue to the advanced development in the last few decades, License Plate Recognition (LPR) systems are quite popular. LPR systems are being developed and implemented by many countries due to the increasing number of traffic violations and crimes, these systems helps in the Automated Ticketing, Automated Parking Management, Toll Station and Traffic Violation and Surveillance.
In the era of advanced technology, everything is changing from the manual system to the automated system but in every organization, we still use the old methods of capturing events manually by using CCTV.
To provide automated and autonomous system for profiling, an integration of image processing with Big Data and AI is required that will automatically record and store the profile of the vehicles coming in the organization. For accessibility a Cloud based system is proposed that will keep the profiling data for tracking.
Project ObjectivesAn automated profiling system is an application which is capable of performing detection, tracking and storage of activity for querying. The objective is to automate the profiling system by integrating the license plate recognition technology with Big Data. In conventional systems such task is performed using CCTV manually which is neither automated nor autonomous. By using this system intelligent queries can be answered with the profiling data. The area comes under the umbrella of Sousveillance.
Project Implementation MethodLicense plate detection technique:
Machine learning is used to detect license plate, and the character will recognize using tensor flow library. After identification of license plate character, the number will store into database with enter or exit time of vehicle. There are following stages of complete license plate detection from capture image to query.
Input Image:
The system captures an image of vehicle from real-time video and converts it into a grayscale image for finding a license plate, because grayscale image will produce a better result for identifying license plate. After this machine learning will start procedure of recognizing license plate.
Processing:
Convert the grayscale image to binary image and apply morphological transformation using OpenCV image library to detect license plate location. It will remove all noises from image, and make it image clear. This filter makes bright the location of license plate. The next step is to apply Gaussian blur filter in binary image to smooth the image. After the implementation of this filter, the image will more bright and clear.
The next step is to remove the background from image which has same color; the only character will show after this. Because we need only license plate, so remove all unnecessary things in image so it will help for better result.
Extract a license plate image from vehicle image, then using segmentation technique; it will make a segment of all characters from image. Now using a trained algorithm, it will detect a character from license plate image and store it into database with vehicle entrance or exit time and date.
Query:
The vehicle license plate profiling data stored in cloud server, the system will quick response of query and give output. ex: show the vehicle between 4pm to 5pm of 25th July 2018. It will show all the vehicle license plate number which is entered and exit in that time.
Benefits of the Project1) Vehicle Tracking.
2) Manageable Records.
3) Reduce labour cost.
4) Time saving.
5) Safety of premises.
6) Secure Entrance Exit.
7) Autonomous Solution.
Extensiblity:
1) Traffic Managent
2) Control Crminal Activities
3) E-Ticketing/E-Chalan
4) Extend to City Level
5) Toll Roads
Technical Details of Final DeliverableThe proposed system will be able to detect the license plate of the vehicle entering in the premises, after detecting the license plate system will automatically store the detected characters in the storage with the timestamp in order to achieve the profiling of the vehicle. The proposed system will also help in answering the complex queries related to the events occur during the time vehicle entered/exit or parked in the premises, if required.
Final Deliverable of the Project Software SystemType of Industry IT , Transportation , Security Technologies Artificial Intelligence(AI), Cloud Infrastructure, Big DataSustainable Development Goals Industry, Innovation and Infrastructure, Sustainable Cities and CommunitiesRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 80000 | |||
| REOLINK C2 Pro Home Security Camera | Equipment | 2 | 13000 | 26000 |
| Nvidia GeForce GTX 1060 Graphics Card | Equipment | 1 | 44000 | 44000 |
| camera cable, wires , remote control device and other detection gadget | Miscellaneous | 1 | 10000 | 10000 |