Number Plate Detector

Published on 02 Jul 2021. Written by Jagmohan

At the entrance of most buildings, the security guard makes a note of vehicle number. It takes a lot of effort and it is a time-consuming process. It may also lead to various security issues. In this project, you will extract the vehicle number using a surveillance camera.

Read more..

You will use the concepts of video processing to extract the vehicle number. Raspberry Pi will act as the brain of the system. You will use Python as the programming language for this project.

This will save a lot of effort and time. By implementing this project, every building will be more secure.

Project Description:

Raspberry Pi 3: It is a third-generation, single-board computer. It contains a 64-bit quad-core processor. The clock speed is 1.5 GHz. An external micro SD card provides memory to Raspberry Pi. You can give instructions using the Raspbian OS.

Raspbian OS: It is similar to Windows or Ubuntu OS. You can code the Raspberry Pi using Python in the Raspbian OS. It is user-friendly and allows multi-tasking.

OpenCV: It is a library that allows image and video processing. You will use it to capture the front side image of the vehicle. Then the image can be altered as per requirement.

Pytesseract: It is an open-source library provided by Google. You will use it to convert the image into text.

Surveillance Camera: Almost all building can use it for their security. You will use it to capture video continuously. It will keep recording the incoming vehicle.

Project implementation:

1. Open Raspbian OS on your system as all the coding is already done.
2. Now start the coding in Python in the Raspbian OS.
3. First, type the code for continuous video capturing.
4. Now convert the video into the required format, i.e., BGR.
5. Find the region of interest. You will use OpenCV to do this. ROI is the part of the image where major operations have to be done. In case of a vehicle, the ROI is always a rectangle, i.e., number plate.
6. Change the perspective of video as per need.
7. Smooth the video to make it ready for various operations.
8. Track the colour of the number plate. This will make it easy to point out the vehicle number.
9. Detect the edges in the video by using various algorithms.
10. Draw the contours to calculate the features of the image.
11. After implementing all advanced OpenCV operations on video, we get the image of the number plate. Finally, use Pytesseract library to convert the extracted image into text.

Programming language: Python

Kit required to develop Number Plate Detector:

• Raspberry Pi
• Surveillance Camera

Technologies you will learn by working on Number Plate Detector:

• Computer Vision
• Smart Camera

Any Questions?