Automatic number plate recognition (ANPR) with Yolov9 and EasyOCR

Automatic Number Plate Recognition (ANPR), also known as License Plate Recognition (LPR), is a technology that uses optical character recognition (OCR) to automatically read and interpret license plates on vehicles.
For queries: You can comment in comment section or you can email me at [e-mail]
Demo video: github.com/AarohiSingla/YOLOv...
ANPR technology is widely used for various purposes, including:
Traffic Management: Monitoring and managing traffic flow and congestion.
Law Enforcement: Identifying and tracking vehicles of interest, such as stolen cars or those associated with criminal activities.
Parking Management: Enforcing parking regulations and managing parking lots.
Toll Collection: Automating toll booths for efficient and fast collection.
Border Control: Monitoring vehicles at border crossings for security purposes.
#computervision #anpr #yolo #yolov9 #yolov8 #objectdetection #pytorch

[@CodeWithAarohi]

Code is available for channel members (Contribution level 2)

[@arnavthakur5409]

Absolutely phenomenal piece of work ma'am.

[@pifordtechnologiespvtltd5698]

Good one

[@Sunil-ez1hx]

Exceptionally very well explained

[@CodeWithAarohi]

Glad it was helpful!

[@rishabhsheoran6959]

Hey Aarohi! Thanks for your videos. I'm updating myself with the latest YOLOs from your channel. Much appreciated effort! All the Best! :D

[@CodeWithAarohi]

Thank you so much for your kind words and support! I'm delighted to hear that you're finding the YOLO tutorials on my channel helpful in keeping yourself updated with the latest advancements. Wishing you all the best in your learning journey! 😊🚀

[@Sans09876]

@@CodeWithAarohi chatgpt reply😏

[@user-wt7bs4ht4h]

mam u r up to the updates in deep learning mam ,this very helpful for those who are interested to learn deeplearning ,i'm glad to have your channel in youtube

[@CodeWithAarohi]

It's my pleasure

[@soravsingla8782]

Keep up the good work 👍👍

[@CodeWithAarohi]

Thanks

[@raghavsharma6430]

thanks for sharing, mam

[@CodeWithAarohi]

Most welcome 😊

[@alexandrelc4361]

Hi , tks for sharing your knowledge , you are very didactic, can you post the github link to all the files.

[@jeyashreej6901]

why after running my weights folder is empty ? where i went wrong mam

[@litziadrianacruz7583]

Hey Aarohi! I love your videos! I'm curious, is it possible to combine yolo for license plate detection and a multimodal to have a description of the license plates?

[@CodeWithAarohi]

Yes, you can do that.

[@biggiewalata]

Thanks

[@CodeWithAarohi]

Thanks

[@AmandeepSingh-uq3wp]

Since you have trained the model with a new dataset, so will it be able to predict the the type of vechiles like car,bus,etc as well along with the license plate at the same time?

[@CodeWithAarohi]

Yes, You can train the model to detect car, bus etc with license plate also

[@joaomartins5069]

Nice work. Does this works well when the car is not right in front of the camera? I had problems with that in my previous projects...

[@CodeWithAarohi]

It depends how clear the number is visible if the vehicle is not in front of camera.

[@archishasingh7299]

Great work, easy to follow Where can I get this sample demo video from?

[@CodeWithAarohi]

github.com/AarohiSingla/YOLOv9/blob/main/a.mp4

[@nurikhsan8969]

How many hours does it take to train this model? There is no output from output cell displayed so i dont know the progress

[@CodeWithAarohi]

ON RTX 3090, it took around 30 minutes

[@jyotpatel1102]

mam can you please guide how to use this model to work directly on live feed or on cctv camera

[@CodeWithAarohi]

Just change the source to 0 in case of webcam and if you want to test on rtsp stream then provide the link of your stream in source parameter.

[@newupdates88]

something went wrong my weights folder is empty, can any one help me

[@danysyauqi6279]

can you gift link repository?

[@chanc31]

Please can you post the github link to all the files

[@mdriad4521]

can you make streamlit web application for this types of project?

[@CodeWithAarohi]

Noted!

[@kunalsingh153]

mam the dataset that you mentioned has been deleted can you provide some other link for it?

[@CodeWithAarohi]

You can pick any other dataset related to license plates from Roboflow universe

[@it_ty_33_nikhilpatil74]

Mam is this possible to Apply ANPR to live video stream ?

[@CodeWithAarohi]

Yes but implementing ANPR on a live video stream can be resource-intensive, especially if you're processing high-resolution video in real-time.

[@IgrisSama]

Thank you for this great video mam. I tried to run the model on my system. Everything works fine until train custom model step. On the prediction with trained model step, it shows me this error AttributeError: 'list' object has no attribute 'device' Edit: Solved this problem. But on last step, another error showed up. AttributeError: 'Annotator' object has no attribute 'draw' Can you help regarding this?

[@CodeWithAarohi]

Open utils/general.py Line no - 903 Replace prediction = prediction[0] with prediction = prediction[0][0]

[@abhinash-bhagat]

Hey @IgrisSama how did you solve the first error "AttributeError: 'list' object has no attribute 'device'" ?

[@IgrisSama]

@@abhinash-bhagat @codewitharohi has given the solution

[@abhinash-bhagat]

@@IgrisSama Yup. I was changing the code in my local files when I was working on Collab. It's solved now after I modified it on drive's file.

[@grayelearning772]

Please share the full code for Automatic Number Plate Recognition (ANPR) project

[@CodeWithAarohi]

Shared on your email

[@amalmirza9823]

Can you please a make video for explainable AI for yolov9..

[@amalmirza9823]

grad cam or eigen cam ..any method

[@CodeWithAarohi]

I will try

[@web3devp]

( yolov9-e_custom.yaml ) doesn't exist in the models/detect file

[@CodeWithAarohi]

you can download it from yolov9 github repo

[@web3devp]

@@CodeWithAarohi so does that mean that i will have to paste the downloaded model and peste it in my model in yolov9 and change from yolov9-e.yaml to yolov9-e-converted.pt in this command { python train_dual.py --workers 8 --batch 8 --img 640 --epochs 200 --data data.yaml --weights ../yolov9-e.pt --device cpu --cfg models/detect/yolov9-e.yaml --hyp data/hyps/hyp.scratch-high.yaml }

[@web3devp]

@@CodeWithAarohi now showing this error urllib.error.URLError:

[@web3devp]

why showing this error when starting training command urllib.error.URLError:

[@CodeWithAarohi]

To resolve it, ensure your system's date and time are correct, update SSL certificates, and verify network configurations.

[@Elmuchacho]

Hello possible live vidéo ocr

[@CodeWithAarohi]

Live video OCR involves capturing video frames from a camera feed in real-time, processing each frame to detect and recognize text using OCR techniques, and then displaying the recognized text on the screen. We can perform that using the code mentioned in this video because we are performing the same steps on test video.

[@kolboy2028]

What GPU you are using?

[@CodeWithAarohi]

Rtx 3090

[@Engineer-Md.AkbarHossain]

source code please

[@aryansingla2037]

My son is need for your help

[@CodeWithAarohi]

You can mail me your query at [e-mail]