Python PDF识别之文本内容与表格提取方法总结

import pdfplumber

pdf_path = "你的PDF文件路径"

with pdfplumber.open(pdf_path) as pdf:
    text = ""
    for page in pdf.pages:
        page_text = page.extract_text()
        text += page_text
        print(page_text)

import fitz

doc = fitz.open(pdf_path)
text = ""
for page in doc:
    page_text = page.get_text()
    text += page_text
    print(page_text)

import PyPDF2

def extract_text_from_pdf(pdf_path):
    pdfReader = PyPDF2.PdfReader(pdf_path)
    text = ""
    for page_num in range(len(pdfReader.pages)):
        page = pdfReader.pages[page_num]
        text += page.extract_text()
    return text

print(extract_text_from_pdf(pdf_path))

from pdfminer.high_level import extract_text

def extract_text_from_pdf(pdf_path):
    text = extract_text(pdf_path)
    return text

print(extract_text_from_pdf(pdf_path))

import pdfquery

def extract_specific_text_from_pdf(pdf_path):
    pdf = pdfquery.PDFQuery(pdf_path)
    pdf.load()
    text = pdf.pq('LTTextLineHorizontal:in_bbox("100, 500, 400, 600")').text()
    return text

print(extract_specific_text_from_pdf(pdf_path))

from pdf2image import convert_from_path
import pytesseract

def extract_text_from_scanned_pdf(pdf_path):
    pages = convert_from_path(pdf_path)
    text = ""
    for page in pages:
        text += pytesseract.image_to_string(page, lang='chi_sim') + "\n"
    return text

print(extract_text_from_scanned_pdf(pdf_path))

from pdf2image import convert_from_path
from paddleocr import PaddleOCR

def extract_text_from_scanned_pdf(pdf_path, lang='ch'):
    ocr = PaddleOCR(use_angle_cls=True, lang=lang)
    pages = convert_from_path(pdf_path)
    text = ""
    for page in pages:
        result = ocr.ocr(page, cls=True)
        for idx in range(len(result)):
            res = result[idx]
            for line in res:
                text += line[1][0] + "\n"
    return text

print(extract_text_from_scanned_pdf(pdf_path))

from aspose.pdf import Document
from aspose.pdf.text import TextFragmentAbsorber

def extract_text_from_pdf(pdf_path):
    pdf_doc = Document(pdf_path)
    absorber = TextFragmentAbsorber()
    text = ""
    for page in pdf_doc.pages:
        page.accept(absorber)
        for fragment in absorber.text_fragments:
            text += fragment.text + "\n"
        absorber.text_fragments.clear()
    return text

print(extract_text_from_pdf(pdf_path))

import textract

def extract_text_from_pdf(pdf_path):
    text = textract.process(pdf_path, method='pdfminer')
    print("-----------textract--------------------")
    print(text.decode('utf-8'))
    return text.decode('utf-8')

print(extract_text_from_pdf(pdf_path))

import pdfplumber

with pdfplumber.open(pdf_path) as pdf:
    for page in pdf.pages:
        table = page.extract_table()
        if table:
            for row in table:
                print(row)

import camelot
import pandas as pd

def extract_tables_from_pdf(pdf_path):
    tables = camelot.read_pdf(pdf_path, flavor='stream')
    combined_df = pd.concat([table.df for table in tables])
    print("-----------camelot--------------------")
    return combined_df

tables_df = extract_tables_from_pdf(pdf_path)
print(tables_df)

from aspose.pdf import Document
from aspose.pdf.text import TextFragmentAbsorber

def extract_tables_from_pdf(pdf_path):
    pdf_doc = Document(pdf_path)
    absorber = TextFragmentAbsorber()
    tables = []
    for page in pdf_doc.pages:
        page.accept(absorber)
        for fragment in absorber.text_fragments:
            # 处理表格数据
            pass
    return tables

print(extract_tables_from_pdf(pdf_path))