CODEWITHSUNDEEP
pythonocrtesseract
Olivier Van D'huynslager
Olivier Van D'huynslager

Reputation: 37

detect and split image for OCR

I am trying to OCR standard forms (they are scanned both front and back)

enter image description here enter image description here

I only want to OCR The second image on the scan (the one with the textual information) - is there a way to detect and split them, and only process the right one? Sorry if I'm missing out on something essential, just starting off. 

 import pytesseract as tess
    import os
    from PIL import Image
    import pandas as pd
    import tesserocr



    path = "/Users/oliviervandhuynslager/PycharmProjects/OCR/DC_SCANS_TEST" ##path to directory (folder) where the images are located

    count = 0
    fileName = [] #create empty list that will contain the original filenames
    fullText = [] #create empty list to store the OCR results per file
    for imageName in os.listdir("/Users/oliviervandhuynslager/PycharmProjects/OCR/DC_SCANS_TEST"):
        count = count + 1
        fileName.append(imageName)
        fileName.sort()#generate list from texts.
    #%%
     # APPEND (OCR) text from images TO LIST fullText
    for imageName in os.listdir("/Users/oliviervandhuynslager/PycharmProjects/OCR/DC_SCANS_TEST"):
        inputPath = os.path.join(path, imageName)
        img = Image.open(inputPath)
        text = tess.image_to_string(img, lang="eng")
        fullText.append(text)

Upvotes: 0

Views: 1515

Answers (1)

user898678
user898678

Reputation: 3328

Here is working example for presented images:

import cv2
import numpy as np
import pytesseract

pytesseract.pytesseract.tesseract_cmd=r"C:\Program Files\Tesseract-OCR\tesseract.exe"

img = cv2.imread("BFezy.png", 0)

kernel = np.ones((25, 25), np.uint8)
eroded = cv2.erode(img, kernel, iterations=2)
dilated = cv2.dilate(eroded, kernel, iterations=1)
thresholded = cv2.threshold(dilated, 150, 255, cv2.THRESH_BINARY_INV)[1]
countours = cv2.findContours(th, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)[0]
if len(countours) == 2:
    x, y, w, h = cv2.boundingRect(countours[0])
    crop = img[y:h + y, x:w + x]
    text = pytesseract.image_to_string(crop)
    print(text)

Upvotes: 2

Related Questions