Oil filling management system for oil Terminals
Project will aim at automating the petrolum compounds and filling stations at scale.We will develop a Computer Vision based solution to montior and authenticate any incoming Oil truck for following: Safety parameters Number Plate of Vehicles These parameters
2025-06-28 16:34:18 - Adil Khan
Oil filling management system for oil Terminals
Project Area of Specialization Artificial IntelligenceProject SummaryProject will aim at automating the petrolum compounds and filling stations at scale.We will develop a Computer Vision based solution to montior and authenticate any incoming Oil truck for following:
- Safety parameters
- Number Plate of Vehicles
These parameters will then be verified from Database and will be displayed on a screen that is fitted outside the main gate.
Similar set of parameters will be stored in the local database and will be displayed on local server.
At Filling station,a camera will detect truck's arrival and will record the time.when truck is done filling the camera will detect and record the exit time.
Project ObjectivesWe will develop an autonomous and secure solution for the company using Aritifical Intelligence(AI) and Internet of things(IOT) technology along side edge deployement technology.
Project Implementation MethodIn this project we will utilize three of cutting edge technologies,
- Computer Vision
- Internet of things
- Edge deployement
At entrance gate,we will use YOLO algorithm to detect and process wether there is a truck at main gate or not.Once a truck is detected it will trigger the detection of safety parameter detection system which will check if the truck has fire extingusher or not.Once it is detected it will trigger the second part of this process i.e photo OCR
We will use Photo OCR model to detect and process the number plate at entrance gate.We will send the processed number plate to local database for authorization and will receive back the following information.
- Driver's Name
- Number Plate
- Authorization status(True/False)
Similar information will be stored in local database alongside time of entry and safety parameters.
At the filling station,a differential flow meter is used to measure when the filling has started and will send the signal to edge device.Edge device will process the number plate to store the following.
- Number Plate
- Start time of filling
- End time of filling
- Filling quantity
In order to explain the benefit we will address the common problems faced by tradational methodology that we aim to address.Current system has too much human involvement which leads to following issues.
- Data Manipulation
- Slower Process
- Calculation error
- human dependent
- Insecure
Our project will lead to following benefits for the company which will help them in both long and short run.
- Reduced time
- Improved efficency
- Database formation
- Data Warehousing
- Security Upgrade
- Digitalization
- Lesser humman involvement.
We will use three cameras and two edge devices to simulate the following environment.
We will use two cameras to simulate the entry gate scenerio of filling station.One camera will be used for vehicle detection and safety parameters detection.Second camera will be used for Photo OCR algorithm for the process of Number plate detection.
Both machine learning models will be deployed using tensorflow lite on raspberry pi.Database that we will use will be sqlite db.
We will use a digital differential flow meter to detect the time filling has started.This time stamp will be stored in database.We will also measure the time filling has ended and will store the timestamp in database sqlite.
We will deploy a tensorflow lite model similar to the one deployed at main gate on preferably jetson tx2 edge device.
Final Deliverable of the Project HW/SW integrated systemCore Industry ITOther Industries Petroleum Core Technology Artificial Intelligence(AI)Other Technologies Internet of Things (IoT)Sustainable Development Goals Industry, Innovation and Infrastructure, Responsible Consumption and ProductionRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 55560 | |||
| Raspberry pi | Equipment | 2 | 10000 | 20000 |
| Cameras | Equipment | 3 | 1020 | 3060 |
| LED screen | Equipment | 1 | 10000 | 10000 |
| adaptors for rpi | Equipment | 2 | 750 | 1500 |
| Casing and fittings. | Equipment | 5 | 750 | 3750 |
| Differential flow meter. | Equipment | 1 | 15000 | 15000 |
| Printing reports. | Miscellaneous | 750 | 3 | 2250 |