Project practice - document scanning OCR recognition

One 、 Complete project code

import numpy as np
import argparse
import cv2

# Set parameters 
ap =argparse.ArgumentParser()
ap.add_argument("-i","--image",required =True,help="Path to the image to be scanned")
args = vars(ap.parse_args())

def order_points(pts):
    # There are four coordinate points 
    rect = np.zeros((4,2),dtype= "float32")

    # Find the corresponding coordinates in order 0123 Namely   Top left , The upper right , The lower right , The lower left 
    # Calculate top left , The lower right 
    s = pts.sum(axis=1)
    rect[0] = pts[np.argmin(s)]
    rect[2] = pts[np.argmax(s)]

    # Calculate top right . The lower left 
    diff = np.diff(pts,axis=1)
    rect[1] = pts[np.argmin(diff)]
    rect[3] = pts[np.argmax(diff)]

    return rect
def four_point_transform(image,pts):
    # Get the input coordinate point 
    rect = order_points(pts)
    (tl,tr,br,bl)=rect

    # Calculate the input w and h value     Under the root x square +y square 
    widthA = np.sqrt(((br[0] - bl[0]) ** 2) + ((br[1]-bl[1]) ** 2))
    widthB = np.sqrt(((tr[0] - tl[0]) ** 2) + ((tr[1] - tl[1]) ** 2))
    maxWidth = max(int(widthA),int(widthB))

    heightA = np.sqrt(((tr[0] - br[0]) ** 2) + ((tr[1] - br[1]) ** 2))
    heightB = np.sqrt(((tl[0] - bl[0]) ** 2) + ((tl[1] - bl[1]) ** 2))
    maxHeight = max(int(heightA), int(heightB))

    # Corresponding coordinate position after transformation 
    dst = np.array([[0, 0], [maxWidth - 1, 0], [maxWidth - 1, maxHeight - 1],
                    [0, maxHeight - 1]], dtype="float32")  #3x3 Matrix 

    # Calculate the transformation matrix 
    M =cv2.getPerspectiveTransform(rect,dst)
    warped =cv2.warpPerspective(image, M ,(maxWidth,maxHeight))

    # Returns the transformed result 
    return warped

def  resize(image, width=None,height=None,inter=cv2.INTER_AREA):
    dim =None
    (h, w)=image.shape[:2]
    if width is None and height is None:
        return image
    if width is None:
        r = height / float(h)
        dim = (int(w * r), height)
    else:
        r = width / float(w)  # The proportion 
        dim = (width, int(h * r))
    resized = cv2.resize(image, dim, interpolation=inter)
    return resized

# Read input 
image = cv2.imread('E:/anaconda3/picture/receipt.jpg')
# The coordinates change the same 
ratio = image.shape[0] / 500.0  
orig = image.copy()

image = resize(orig, height =500)

# Preprocessing 
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY) # grayscale 
gray = cv2.GaussianBlur(gray, (5,5),0)   # Gauss filtering , Used to remove some noise points 
edged = cv2.Canny(gray, 75, 200)    # Edge detection , In order to extract the contour of the back 

# Display preprocessing results 
print("STEP 1:  edge detection ")
cv2.imshow("Image", image)
cv2.imshow("Edged", edged)
cv2.waitKey(0)
cv2.destroyAllWindows()

#  Contour detection 
cnts = cv2.findContours(edged.copy(), cv2.RETR_LIST, cv2.CHAIN_APPROX_SIMPLE)[1] 
cnts = sorted(cnts, key = cv2.contourArea, reverse = True)[:5]  

#  Traverse the outline 
for c in cnts:
    #  Calculate the contour approximation 
    peri = cv2.arcLength(c, True)  # Contour length 
    # C Represents the set of input points 
    # epsilon Represents the maximum distance from the original contour to the approximate contour , It's an accuracy parameter 
    # True It means closed 
    approx = cv2.approxPolyDP(c, 0.02 * peri, True)  

    # 4 Take it out at one point 
    if len(approx) == 4:  # Get an approximation  4 A little bit 
        screenCnt = approx   # Approximate results 
        break  # Take it out 

#  Display the results 
print("STEP 2:  Get profile ")
cv2.drawContours(image, [screenCnt], -1, (0, 255, 0), 2)   # Draw the outline 
cv2.imshow("Outline", image)
cv2.waitKey(0)
cv2.destroyAllWindows()

#  Perspective transformation  
warped = four_point_transform(orig, screenCnt.reshape(4, 2) * ratio) 

#  Binary processing 
warped = cv2.cvtColor(warped, cv2.COLOR_BGR2GRAY)
ref = cv2.threshold(warped, 100, 255, cv2.THRESH_BINARY)[1]
cv2.imwrite('scan.jpg', ref)
#  Display the results 
print("STEP 3:  Transformation ")
cv2.imshow("Original", resize(orig, height=650))
cv2.imshow("Scanned", resize(ref, height=650))
cv2.waitKey(0)

Two 、 Specific operation

1、 First import the toolkit , Set parameters ( Set picture reading path )

import numpy as np
import argparse
import cv2

# Set parameters 
ap =argparse.ArgumentParser()
ap.add_argument("-i","--image",required =True,help="Path to the image to be scanned")
args = vars(ap.parse_args())

2、 Read input , After image reading , Calculate its four coordinate points , To do edge detection, it is necessary to have the same change in coordinates . then , Then preprocess the pictures （ Go to grayscale 、 Gauss filtering - In order to filter out some noise points 、 edge detection ）

# Read input 
image = cv2.imread('E:/anaconda3/picture/receipt.jpg')
# The coordinates change the same 
ratio = image.shape[0] / 500.0  
orig = image.copy()

image = resize(orig, height =500)

# Preprocessing 
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY) # grayscale 
gray = cv2.GaussianBlur(gray, (5,5),0)   # Gauss filtering , Used to remove some noise points 
edged = cv2.Canny(gray, 75, 200)    # Edge detection , In order to extract the contour of the back 

# Display preprocessing results 
print("STEP 1:  edge detection ")
cv2.imshow("Image", image)
cv2.imshow("Edged", edged)
cv2.waitKey(0)
cv2.destroyAllWindows()

Output results :

After edge detection , Carry out specific contour detection

3、 Contour detection

cnts = cv2.findContours(edged.copy(), cv2.RETR_LIST, cv2.CHAIN_APPROX_SIMPLE)[1]  #
cnts = sorted(cnts, key = cv2.contourArea, reverse = True)[:5]   

 Will detect all the contours on this picture , But what we need is the outermost , So in order to calculate his external things , So sort the top five largest outlines , Then traverse each contour :

for c in cnts:
    #  Calculate the contour approximation 
    peri = cv2.arcLength(c, True)  # Contour length 
    # C Represents the set of input points 
    # epsilon Represents the maximum distance from the original contour to the approximate contour , It's an accuracy parameter 
    # True It means closed 
    approx = cv2.approxPolyDP(c, 0.02 * peri, True)  # The contour points are 【 The approximate 】（c Is every outline /0.02 * peri Namely epsilon/）

    # 4 Take it out at one point 
    if len(approx) == 4:  # Get an approximation  4 A little bit 
        screenCnt = approx   # Approximate results 
        break  # Take it out 

Output results :

thus , The text in the picture highlights , To get the specific text content in the picture, you need to install tesseract_ocr

3、 ... and 、tesseract_OCR Installation and configuration

1、 Website of installation package ：https://digi.bib.uni-mannheim.de/tesseract/

Select the latest version to download , Set up the installation path （ I am a E:\opencv)

2、 Configure environment variables ： attribute > Advanced settings > environment variable > In the upper and lower boxes path Add it all in E:\opencv The path of Just OK

3、WIN+R Input cmd Post input tesseract -v To test , Get the version number and you will succeed

4、tesseract XXX.png Get the results   Enter a picture

5、tesseract XXX.png   result  Get the text result recognized by the picture    This means that tesseract Installation successful

6、 On the command bar tesseract In the directory of pip install pytesseract, install pytesseract

7、 stay anaconda3> lib> site-packges> pytesseract.py >tesseract_cmd It is amended as follows Absolute path that will do

  The absolute path is tesseract Specific installation path .

Come here tesseract You can run to recognize the image text content :

from PIL import Image
import pytesseract
import cv2
import os

preprocess = 'blur' #thresh

image = cv2.imread('scan.jpg')
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)

if preprocess == "thresh":
    gray = cv2.threshold(gray, 0, 255,cv2.THRESH_BINARY | cv2.THRESH_OTSU)[1]

if preprocess == "blur":
    gray = cv2.medianBlur(gray, 3)
    
filename = "{}.png".format(os.getpid())
cv2.imwrite(filename, gray)
    
text = pytesseract.image_to_string(Image.open(filename))
print(text)
os.remove(filename)

cv2.imshow("Image", image)
cv2.imshow("Output", gray)
cv2.waitKey(0)   

  Will get " The words on the supermarket piece :"

Output :