👉Created a model that will detect a car in a live stream or video and recognize characters on number plate of the car .
👉Secondly , it will use the characters and fetches the owners information using RTO API’s .
👉Created a Web portal where all this information will be displayed (using html,css,and js)
I have seperated the solution into multiple steps.
Step1 - Train car detection model and crop the car image only.
Step2 - The cropped image will send to the car_number_plate_detecton_model where the visible number_plate will be cropped out.
Step3 - Then the cropped image of number plate will be send to Number-detecton model to read the numbera and alphabets
FRONT-END
The front-End of the web_app will look similar to the below image. The user can upload their image using the upload button and then the whole file will be uplaoded to the server. To make it easy I have used s3 to store the image so that it can be downloaded from anywhere.
For Securtiy the web_server only have access to the s3.
The number plate of the user will then be downloaded in the server after which the character detection model will come into play.
As you can see on the right side we have the characters detected as weel as their accuracy!
BEHIND THE SCENES
Car Number plate detection and Extraction
A demo of functioning model.
click on it -> car_number_Plate_detection
As you can see the number plates are detected, cropped and then seperated
character Recognition model training
Now that we have trained the character recognition model we are going to use this to extract the characters present in the number plate
Thank You