Detect and Recognize Car License Plate from a video in real time in OpenCV

Detecting and recognizing a car license plate from a video in real-time involves multiple steps, and can be approached using various methods. In this tutorial, I'll guide you through a basic approach using OpenCV and Python:

Car License Plate Detection and Recognition using OpenCV

1. Setup

Install the necessary libraries if you haven't already:

pip install opencv-python

2. License Plate Detection

For detecting license plates, you can use a Haar cascade classifier:

import cv2

# Load the pre-trained Haar cascades from OpenCV
plate_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_russian_plate_number.xml')

cap = cv2.VideoCapture('path_to_video.mp4')

while cap.isOpened():
    ret, frame = cap.read()
    if not ret:
        break

    gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
    plates = plate_cascade.detectMultiScale(gray, 1.1, 10)

    for (x, y, w, h) in plates:
        cv2.rectangle(frame, (x, y), (x + w, y + h), (0, 255, 0), 2)
        roi_gray = gray[y:y + h, x:x + w]
        roi_color = frame[y:y + h, x:x + w]

        # Here you can add the recognition part

    cv2.imshow('Video', frame)
    if cv2.waitKey(1) & 0xFF == ord('q'):
        break

cap.release()
cv2.destroyAllWindows()

3. License Plate Recognition

License plate recognition is more complex and typically involves Optical Character Recognition (OCR). A popular tool for this is Tesseract. You can install it via pip:

pip install pytesseract

Make sure you also have Tesseract installed on your system.

After setting up Tesseract, you can recognize license plates as follows:

import pytesseract

# ...

# Inside the for loop, after detecting the license plate:
        text = pytesseract.image_to_string(roi_gray, config='--psm 8')
        cv2.putText(frame, text, (x, y - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.8, (0, 0, 255), 2)

# ...

This code captures license plates from a video and displays them in real-time. Adjust the parameters of detectMultiScale and the configurations of Tesseract as needed to improve accuracy.

Notes:

This tutorial provides a basic approach to license plate detection and recognition. For more robust results, consider using more advanced methods like deep learning-based models.
Real-time processing speed may vary based on the efficiency of the methods used and the processing capabilities of your machine. If real-time performance is essential, you might need to optimize the pipeline or employ faster algorithms.

Car license plate recognition with OpenCV in Python:

License plate recognition involves detecting and recognizing license plates on vehicles. OpenCV provides tools for image processing and computer vision, making it suitable for license plate recognition tasks.

import cv2

# Load the pre-trained license plate classifier
plate_cascade = cv2.CascadeClassifier('haarcascade_license_plate.xml')

# Read an image
img = cv2.imread('car_image.jpg')

# Convert to grayscale
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

# Detect license plates in the image
plates = plate_cascade.detectMultiScale(gray, scaleFactor=1.3, minNeighbors=5)

# Draw rectangles around the license plates
for (x, y, w, h) in plates:
    cv2.rectangle(img, (x, y), (x+w, y+h), (255, 0, 0), 2)

# Display the image with detected license plates
cv2.imshow('License Plates', img)
cv2.waitKey(0)
cv2.destroyAllWindows()

Real-time license plate detection and recognition using OpenCV:

Extending license plate recognition to real-time scenarios involves processing video frames continuously and applying license plate detection and recognition.

import cv2

# Load pre-trained license plate classifier and OCR engine
plate_cascade = cv2.CascadeClassifier('haarcascade_license_plate.xml')
ocr = cv2.text.OCRTesseract_create()

# Open a connection to the webcam (0 represents the default camera)
cap = cv2.VideoCapture(0)

while True:
    # Read a frame from the webcam
    ret, frame = cap.read()

    # Convert to grayscale
    gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)

    # Detect license plates in the frame
    plates = plate_cascade.detectMultiScale(gray, scaleFactor=1.3, minNeighbors=5)

    # Recognize license plates using OCR
    for (x, y, w, h) in plates:
        plate_roi = frame[y:y+h, x:x+w]
        plate_text = ocr.run(plate_roi)
        cv2.putText(frame, plate_text, (x, y-10), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 0), 2)

    # Display the frame with license plate recognition
    cv2.imshow('License Plate Recognition', frame)

    # Break the loop if 'q' key is pressed
    if cv2.waitKey(1) & 0xFF == ord('q'):
        break

# Release the webcam and close the window
cap.release()
cv2.destroyAllWindows()