pip install pymupdf -i http://pypi.douban.com/simple --trusted-host pypi.douban.com

链接：https://pan.baidu.com/s/18d5X9O2BF8dm3gZI8p0J2w 
提取码：ob96 

import os
import time

import fitz
import glob

import pytesseract
import tesserocr
from PIL import Image
from aip import AipOcr

class PdfHandle():
    #初始化
    def __init__(self):
        self.APP_ID = ''
        self.API_KEY = ''
        self.SECRET_KEY = '9HddR7LxTpEe3zhRTTU41DwbSKEOkQN0'
        self.img_path = './pdf/'
        self.imgs_list = []

    #pdf提取图片
    def fun1(self,filename):
        name = glob.glob(filename)[0]
        doc = fitz.open(name)
        for pg in range(0,doc.pageCount):
            page = doc[pg]
            zoom = int(1000)
            rotate = int(0)
            trans = fitz.Matrix(zoom / 100.0 ,zoom / 100.0).preRotate(rotate)
            pm = page.getPixmap(matrix=trans,alpha=0)
            pm.writePNG(self.img_path + '{}.png'.format(str(pg)))

    #调用百度api提取文字
    def tiqu(self):
        self.traverse(self.img_path)
        aip_orc = AipOcr(self.APP_ID, self.API_KEY, self.SECRET_KEY)

        for file_path in self.imgs_list[0:1]:
            file_path = file_path[0]
            options = {}
            options['detect_direction'] = 'true'
            options['probability'] = 'true'
            id_card_side = 'front'
            result = aip_orc.basicAccurate(self.get_file_content(file_path), options)
            # print(result)
            for temp in result['words_result']:
                print(temp['words'])

    #调用百度api提取文字
    def tiqu2(self):
        self.traverse(self.img_path)
        aip_orc = AipOcr(self.APP_ID, self.API_KEY, self.SECRET_KEY)

        for file_path in self.imgs_list:
            file_path = file_path[0]
            print(file_path)
            options = {}
            options["language_type"] = "CHN_ENG"
            options["detect_direction"] = "true"
            options["detect_language"] = "true"
            options["probability"] = "true"
            while 1:
                try:

                    result = aip_orc.basicGeneral(self.get_file_content(file_path), options)
                    break
                except Exception as e:
                    time.sleep(2)

                    continue

            # print(result)
            for temp in result['words_result']:
                print(temp['words'])
            print('=========')

    #读取图片二进制内容
    def get_file_content(self,file_path):
        with open(file_path, 'rb') as f:
            return f.read()

    #遍历目录下的文件图片
    def traverse(self,d):
        dirs = os.listdir(d)  # 获取当前目录的文件列表

        for item in dirs:
            absPath = os.path.join(d, item)  # 拼接绝对路径
            if os.path.isdir(absPath):  # 判断是否是目录
                self.traverse(absPath)  # 递归调用函数
            else:
                if os.path.basename(absPath).endswith('.png'):
                    self.imgs_list.append((absPath,os.path.basename(absPath)))
    #使用 tesseract 识别文字
    def my_tesseract(self):
        img = Image.open(r'D:\bfy\my_code\task_3\pdf\6.png')
        print(pytesseract.image_to_string(img,lang='chi_sim'))


if __name__ == '__main__':
    pdf = PdfHandle()
    # pdf.my_tesseract()
    pdf.tiqu2()