使用Python和OCR进行文档解析的完整代码演示（附代码）

来源：DeepHub IMBA
本文约2300字，建议阅读5分钟
本文中将使用Python演示如何解析文档(如pdf)并提取文本，图形，表格等信息。

https://s2.q4cdn.com/470004039/files/doc_financials/2021/q4/_10-K-2021-(As-Filed).pdf

环境设置

• 以文本方式处理文档:用PyPDF2提取文本，用Camelot或TabulaPy提取表，用PyMuPDF提取图形。
• 将文档转换为图像(OCR):使用pdf2image进行转换，使用PyTesseract以及许多其他的库提取数据，或者只使用LayoutParser。

# with pippip install python-poppler# with condaconda install -c conda-forge poppler

# READ AS IMAGEimport pdf2imagedoc = pdf2image.convert_from_path("doc_apple.pdf")len(doc) #<-- check num pagesdoc[0]   #<-- visualize a page

# Save imgsimport osfolder = "doc"if folder not in os.listdir():  os.makedirs(folder)p = 1for page in doc:  image_name = "page_"+str(p)+".jpg"    page.save(os.path.join(folder, image_name), "JPEG")  p = p+1

pip install layoutparser torchvision && pip install "git+https://github.com/facebookresearch/detectron2.git@v0.5#egg=detectron2"

pip install "layoutparser[ocr]"

import layoutparser as lpimport cv2import numpy as npimport ioimport pandas as pdimport matplotlib.pyplot as plt

检测

## load pre-trained modelmodel = lp.Detectron2LayoutModel(  "lp://PubLayNet/mask_rcnn_X_101_32x8d_FPN_3x/config",  extra_config=["MODEL.ROI_HEADS.SCORE_THRESH_TEST", 0.8],  label_map={0:"Text", 1:"Title", 2:"List", 3:"Table", 4:"Figure"})
## turn img into arrayi = 21img = np.asarray(doc[i])
## predictdetected = model.detect(img)
## plotlp.draw_box(img, detected, box_width=5, box_alpha=0.2,          show_element_type=True)

## sortnew_detected = detected.sort(key=lambda x: x.coordinates[1])## assign idsdetected = lp.Layout([block.set(id=idx) for idx,block in                    enumerate(new_detected)])## checkfor block in detected:  print("---", str(block.id)+":", block.type, "---")  print(block, end='\n\n')

提取

'''{'0-Title': '...','1-Text': '...','2-Figure': array([[ [0,0,0], ...]]),'3-Table': pd.DataFrame,}'''def parse_doc(dic):  for k,v in dic.items():      if "Title" in k:          print('\x1b[1;31m'+ v +'\x1b[0m')      elif "Figure" in k:          plt.figure(figsize=(10,5))          plt.imshow(v)          plt.show()      else:          print(v)      print(" ")

# load modelmodel = lp.TesseractAgent(languages='eng')dic_predicted = {}for block in [block for block in detected if block.type in ["Title","Text"]]:  ## segmentation  segmented = block.pad(left=15, right=15, top=5,              bottom=5).crop_image(img)  ## extraction  extracted = model.detect(segmented)  ## save  dic_predicted[str(block.id)+"-"+block.type] =                extracted.replace('\n',' ').strip()
# checkparse_doc(dic_predicted)

for block in [block for block in detected if block.type == "Figure"]:  ## segmentation  segmented = block.pad(left=15, right=15, top=5,                        bottom=5).crop_image(img)  ## save  dic_predicted[str(block.id)+"-"+block.type] = segmented
# checkparse_doc(dic_predicted)

for block in [block for block in detected if block.type == "Table"]:  ## segmentation  segmented = block.pad(left=15, right=15, top=5,              bottom=5).crop_image(img)  ## extraction  extracted = model.detect(segmented)  ## save  dic_predicted[str(block.id)+"-"+block.type] = pd.read_csv(                io.StringIO(extracted) )# checkparse_doc(dic_predicted)

import tabulatables = tabula.read_pdf("doc_apple.pdf", pages=i+1)tables[0]

总结

本文的源代码：

https://github.com/mdipietro09/DataScience_ArtificialIntelligence_Utils/blob/master/computer_vision/example_ocr_parsing.ipynb

如果你安装Tesseract有问题的话，请看这个帖子

https://stackoverflow.com/questions/50951955/pytesseract-tesseractnotfound-error-tesseract-is-not-installed-or-its-not-i

编辑：于腾凯