final_car_plate_number_recognition.py

# -- coding: utf-8 --
"""FINAL CAR PLATE NUMBER RECOGNITION

Automatically generated by Colaboratory.

Original file is located at
    https://colab.research.google.com/drive/1dWyreN1962LUNGVNbHY4Be1LbbtKwVa_

# Car plate number Recognition Model
"""

# !nvcc --version ## 3 7 10 11 13 14 17 22 23 24 32 33 41

# !pip install easyocr
# !pip install imutils
# !pip install opencv-python-headless==4.1.2.30
# !pip3 install torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cu113


import cv2
import easyocr

text = ""
# Load image
image = cv2.imread('car.jpeg')
print("hello")
# Convert to grayscale
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)

# Initialize EasyOCR reader
reader = easyocr.Reader(['en'])
print("hry")
# Detect license plate using EasyOCR
results = reader.readtext(gray, detail=0)
for result in results:
    if len(result) == 10 and result.isalnum() and (not result.isnumeric()) and (not result.isalpha()):
        text = result
        print(result)


# Display result
# cv2.imshow('Result', image)
# cv2.waitKey(0)
# cv2.destroyAllWindows()

# import cv2
# import easyocr

# # Load image
# image = cv2.imread(
#     'C:/Users/jahna/Dropbox/My PC (DESKTOP-UKVOUSD)/Desktop/cb/carplatenumber3.jpg')

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

# # Initialize EasyOCR reader
# reader = easyocr.Reader(['en'])

# # Detect license plate using EasyOCR
# results = reader.readtext(gray, detail=0)
# for result in results:
#     if len(result) == 7 and result.isalnum():
#         print(f"License Plate Detected: {result}")
#         # Draw bounding box around license plate
#         coords = reader.recognize(gray, detection_type='textline')
#         x_min, y_min = coords[0][0][0], coords[0][0][1]
#         x_max, y_max = coords[0][2][0], coords[0][2][1]
#         cv2.rectangle(image, (x_min, y_min), (x_max, y_max), (0, 255, 0), 2)
#         break

# # Display result
# cv2.imshow('Result', image)
# cv2.waitKey(0)
# cv2.destroyAllWindows()
# ------------------------
# import cv2
# from matplotlib import pyplot as plt
# import numpy as np
# import imutils
# import easyocr
# # import main
# # from main import *
# import os
# import streamlit as st
# os.environ["KMP_DUPLICATE_LIB_OK"] = "TRUE"

# # preprocessing
# img = cv2.imread(
#     r'C:/Users/jahna/Dropbox/My PC (DESKTOP-UKVOUSD)/Desktop/cb/carplatenumber3.jpg')
# # due red green blue, pixel increase so that is why convert to gray

# gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
# # plt.imshow(cv2.cvtColor(gray, cv2.COLOR_BGR2RGB))  # savetime

# """*Apply Filters and edge detection*"""

# bfilter = cv2.bilateralFilter(gray, 11, 17, 17)  # noise reduced
# edged = cv2.Canny(bfilter, 30, 200)  # DETECT EDGE
# # plt.imshow(cv2.cvtColor(edged, cv2.COLOR_BGR2RGB))
# # when we converted color->gray a lot of data/pixels got reduced due to which the analysis took less time
# # now that we have used bfliter so that more time is reduced

# """*Find contours and apply mask*"""

# # contours stroes polygons in a pic seperatly we can easily detect no. plate
# keypoints = cv2.findContours(
#     edged.copy(), cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
# contours = imutils.grab_contours(keypoints)
# contours = sorted(contours, key=cv2.contourArea, reverse=True)[:10]
# # the sorting in descending order so that better shapes come first
# # chain_approx line is used to simplofy the data ie if we dont use it then even pixels of line would get stored but we need proper polygons

# location = None
# for contour in contours:
#     # how many approx values if we increase value the result would be more rough accordingly
#     approx = cv2.approxPolyDP(contour, 10, True)
#     if len(approx) == 4:
#         location = approx
#         break

# location

# # will finally mask
# mask = np.zeros(gray.shape, np.uint8)
# new_image = cv2.drawContours(mask, [location], 0, 255, -1)
# new_image = cv2.bitwise_and(img, img, mask=mask)

# # plt.imshow(cv2.cvtColor(new_image, cv2.COLOR_BGR2RGB))

# (x, y) = np.where(mask == 255)
# (x1, y1) = (np.min(x), np.min(y))
# (x2, y2) = (np.max(x), np.max(y))
# cropped_image = gray[x1:x2+1, y1:y2+1]

# # plt.imshow(cv2.cvtColor(cropped_image, cv2.COLOR_BGR2RGB))

# """*Easy OCR use*"""

# reader = easyocr.Reader(['en'])  # english
# result = reader.readtext(cropped_image)
# result

# """*Plot the Model*"""

# # if (len(result) == 1):
# #     text = result[0][-2]
# # else:
# #     text = result[1][-2]
# # print(text)
# # print(result, "juuuuu")
# if (len(result) == 1):
#     text = result[0][-2]
# else:
#     text = result[1][-2]
# # text = result[0][-2]
# st.write(text)
# font = cv2.FONT_HERSHEY_SIMPLEX
# # res=cv2.putText(img,text=text, org=(approx[0][0][0], approx[1][0][1]+60), fontFace=font, fontScale=1, color=(0,255,0), thickness=2, lineType=1)
# res = cv2.rectangle(img, tuple(approx[0][0]), tuple(
#     approx[2][0]), (0, 255, 0), 3)
# img = cv2.putText(img, text, tuple(
#     approx[0][0]), font, 0.5, (0, 255, 0), 2, cv2.LINE_AA)
# # plt.imshow(cv2.cvtColor(res, cv2.COLOR_BGR2RGB))