Case study

ANPR-based Automated Payment Solution as a Compelling Differentiator for Brands

We’ve developed a contactless automated payment system that can give a competitive edge to merchants and service providers

Key features

  • Address consumer needs

    Address consumer needs

  • Provide zero-click payments

    Provide zero-click payments

  • Eliminate the need for cash, cards, POS terminals

    Eliminate the need for cash, cards, POS terminals

Industry:
FinTech, Automotive
Market:
Global
Team size:
10 engineers
Project duration:
9 months
Technologies:

AngularJS / Azure AD / Java / KNN / Objective-C / OpenCV

Business challenge

The more disruption technology undergoes, the more tech-savvy customers get. Intellias has been delivering technologically advanced solutions for 18 years, and throughout this time we’ve been watching how continuous innovations in consumer electronics, smartphones, payment systems, and mobility concepts raise the expectations of the end user. In response to increasing digitalization in our daily lives, customers demand easy and flawless interactions with service providers. Optimizing the user experience has never been this crucial in any industry.

With this in mind, global retailers, drive-through restaurants, gas stations, and many other service providers are moving to automate the customer experience to the greatest possible extent. An automated payment system backed by artificial intelligence (AI) can address a lot of pain points for drivers, such as:

  • slow lines at the drive-through
  • lost or misplaced parking tickets
  • reaching for the wallet while at the wheel
  • the need to open the car window to pay via NFC or, what’s worse, to pass cash/coins that might even be dropped
  • hardly readable QR codes

The Intellias team has lots of expertise developing software for the automotive industry. Keeping drivers’ pain points in mind, our team developed an automated payment solution as an internal R&D project.
ANPR-based Automated Payment Solution as a Compelling Differentiator for Brands

Solution

Our R&D project is based on the idea of providing drivers with a unique user experience: zero-click payments for products and services based on a combination of computer vision and geofenced car location tracking.

A vehicle is identified the moment it pulls up to the gas station, parking space, carwash, or any other service center. The next step is trajectory tracking to define the precise gas pump, parking space, or washing station used. Then the license plate is recognized and a payment is made automatically.

The payment system itself is based on a combination of authorization and authentication methods including Automatic Number Plate Recognition (ANPR). ANPR is widely used as an authentication method for free-flow toll roads around the world, so it can be used as the main authentication method even in highly sensitive scenarios such as for payments.

The key technology behind this solution is a combination of image recognition with a mobile app on a user’s smartphone for authorization and a cloud platform that supports image recognition in general and, in particular, Optical Character Recognition (OCR). This combination eliminates the need for installing hardware other than IP cameras at service points. Installed cameras detect number plates with NVR, HD RAID, and OpenALPR and stream information to a computer vision service.

Once the object detection analysis outputs a positive result, it triggers optical character recognition for the license plate detection workflow:

  • Convert the image to grayscale
  • Optimize the image quality with smoothing filters
  • Extract all contours in the image using OpenCV functionality
  • Check the similarity of all contours with predefined symbols using a k-nearest neighbors (KNN) algorithm
  • Use heuristics to reconstruct license plate numbers (e.g. the standard format for a British number plate introduced in 2001 is 2 letters – 2 digits – 3 letters)

A significant benefit of this solution is that it can run even on basic hardware because the KNN algorithm substitutes a neural network for number plate recognition. We trained the algorithm on a data set of Ukrainian license plates and managed to achieve 94.3% recognition accuracy even at long distances, in poor weather conditions, and for dusty license plates.

Business value

Adopting automated payments is an obvious step for intrepid brands that aim to discover more ways of addressing customers’ needs. Places where ANPR-based payments may be used include:

  • gas stations
  • drive-through restaurants
  • retail stores (for pickup)
  • parking lots
  • toll roads
  • car washes
  • car maintenance centers

This technology eliminates the need for cash, cards, POS terminals, and QR codes. Customers can authorize via a smartphone app by entering a license plate number, connecting a card to it, and selecting whether to turn on two-factor authentication.

A zero-click payment option powered by artificial intelligence development provides businesses with an opportunity to increase revenue by decreasing service time and therefore being able to serve more customers. But most importantly, drivers can benefit from this technology by saving time, effort, and money.

All in all, automated payments powered by AI are a win-win solution.

Tell us about your project

I give consent to the processing of my personal data given in the contact form above under the terms and conditions of Intellias Privacy Policy. I want to receive commercial communications and marketing information from Intellias by electronic means of communication (including telephone and e-mail).
* I give consent to the processing of my personal data given in the contact form above under the terms and conditions of Intellias Privacy Policy.

Awards and recognition

logo
logo
logo
logo
logo
logo

Thank you for your message.
We will get back to you shortly.

Thank you for your message.
We will get back to you shortly.