Executive Summary

The primary goal of this project is to enhance security and efficiency at petrol stations by using an LPR (License Plate Recognition) system to automatically recognize and verify number plates before authorizing fuel transactions.

This system aims to prevent fraud and ensure only registeredvehicles access services. With forecourt crime costing UK retailers an average of £10,500 per site annually, the need for such systems is underscored by the effectiveness of organizations like BOSS (British Oil Security Syndicate), which help reduce these losses significantly. Despite a 5.9% reduction in unpaid fuel incidents in Q4 2023, which brought the Forecourt Crime Index back to its Q3 2022 level, the severity of incidents has escalated, as indicated by increased average litters drawn per incident(Source: Ref 1)

Solution

Our system offers robust and reliable vehicle identification and verification, by leveraging state-of-the-art AI models and advanced computer vision techniques, This ensures that all detected vehicles are correctly matched with their registered details from DVLA, providing an additional layer of security

Architecture

Object Detection Architecture

API Architecture

Use case :

Automatic number plate recognition has a lot of use cases such as surveillance, parking management, toll collection, logistics, etc.,