Today, I wanted to talk about a project that we recently worked on at Pet Funeral Services called the Gate Keeper Project. This project was aimed at creating an Automatic License Plate Recognition system using a Raspberry PI and an advanced camera module with a 35mm lens adapter for a 300 mm lens. We coded the system in Python using Flask for intranet server broadcasting and Open ALPR for image recognition, which in turn utilized Open CV.

The hardware was put together by a custom 3D print mount designed in Autodesk Fusion 360. The project was conceptualized as an experiment to set up a camera recognition system to alert the office when customers have arrived so that we could come and greet them at the door. As our customers for the pet cemetery are often in great emotional distress due to their recent loss, our business tries its best to support families as much as possible in this moment.

One of the key features of this project was the camera’s very far reach, as it was able to recognize cars 120m away. This was implemented due to a lack of electricity availability and internet connection near the far gate of the property. Additionally, the project was developed at a relatively low cost, without subscriptions or high-cost camera equipment prices.

We were also very mindful of data privacy policies during the development of this project. As it turns out, license plates are public and do not conflict with GDPR regulations, as what is protected is personal data. Nevertheless, we did consult a legal advisor and set up a notice during the project period.

While the project worked well for a short period during the pandemic, we encountered some challenges with differentiating between Funeral Services customers and Coffee Shop customers who operate in the same premises. As a result, the system would beep for every single car arriving on the property, causing some disruption in the office.

To overcome this challenge, we plan to implement a solution that further integrates the project with our CRM system. This would allow us to retrieve customer information and filter the relevant data. Overall, we are very excited about the potential of this project and are looking forward to exploring further implementation in the future.