使用 OCR 和 OpenCV 仅提取特定信息

Question

我正在尝试从账单中获取特定信息。到目前为止，我一直使用 ocr 和 OpenCV，结果如下：

import cv2 
import pytesseract
import numpy as np

image = cv2.imread('1.png')

# get grayscale image
def get_grayscale(image):
return cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)

# noise removal
def remove_noise(image):
return cv2.medianBlur(image,5)

#thresholding
def thresholding(image):
return cv2.threshold(image, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)[1]

#dilation
def dilate(image):
kernel = np.ones((5,5),np.uint8)
return cv2.dilate(image, kernel, iterations = 1)

#erosion
def erode(image):
kernel = np.ones((5,5),np.uint8)
return cv2.erode(image, kernel, iterations = 1)

#opening - erosion followed by dilation
def opening(image):
kernel = np.ones((5,5),np.uint8)
return cv2.morphologyEx(image, cv2.MORPH_OPEN, kernel)

#canny edge detection
def canny(image):
return cv2.Canny(image, 100, 200)

#skew correction
def deskew(image):
coords = np.column_stack(np.where(image > 0))
angle = cv2.minAreaRect(coords)[-1]
if angle < -45:
    angle = -(90 + angle)
else:
    angle = -angle
(h, w) = image.shape[:2]
center = (w // 2, h // 2)
M = cv2.getRotationMatrix2D(center, angle, 1.0)
rotated = cv2.warpAffine(image, M, (w, h), flags=cv2.INTER_CUBIC, borderMode=cv2.BORDER_REPLICATE)
return rotated

#template matching
def match_template(image, template):
return cv2.matchTemplate(image, template, cv2.TM_CCOEFF_NORMED) 


gray = get_grayscale(image)
thresh = thresholding(gray)
opening = opening(gray)
canny = canny(gray)
cv2.imshow('res', gray)
cv2.waitKey(0)
cv2.destroyAllWindows()

# Adding custom options
pytesseract.pytesseract.tesseract_cmd = r'C:\Program Files\Tesseract-OCR\tesseract.exe'
custom_config = r'--oem 3 --psm 6'
pytesseract.image_to_string(gray, config=custom_config)

我得到的输出是

Out[9]: 'aso en bosaanes sosesoen\nSee arr ee\n[internationale Spedition “works carrier:\nree Meese 
Eaton oro\nSE Eesn Srey alata ascea\ntay See eae ror\nTBlaecaseew £2 saserzaz9gn [acs Sue Saeeats 
Arve\noricore toptetschlBve ta\nbares eye creat tere\nLene et aan Ease\ncoon soos\nreaee\nbenenter 
petachand AiG & co. x8\nese See ete Fests\nsee Sse\npearson | |\nen 7\nFeanséurt an main bawegoansn 
|\npe |\nsor per tantace e/ear0003537\nEl = T=] | = [== |\nSta psa a4 fonstsanern\nLerper 
atcnen\nwe\n20 ocd hoes ale 22ers wf\n30 ped londed on pwc aoasonnr #0\n35 ped londed on pwc 2008es00 
#0\n64 pcs loaded on| PMC BO3BBART MD &\n[ental — |\n=\n|\nSJ |] Spscrinan copnapen as wtshan momen 
ante\nart veins otetrich cata 60. RAS sem\n[re ote\n[\\gesoago |__| tars ena Detrich ea\nTon anine 
Setrion cn a co. eta a5 scan\nSS aan ee ee\nee eS] - 
esemen\ncision\n\x0c'

我只需要名称、送货地址、数量等具体信息，而不需要所有字符。此外，输出全部混合。谁能帮我解决这个问题？任何代码或任何其他帮助将不胜感激。

Answer 1

您可以使用pytesseract.image_to_pdf_or_hocr()，选择hocr 作为输出格式。这将包含字符、单词和行级别的边界框。