python调阿里OCR_增值税发票，定额发票，混贴发票，表格_识别，处理数据写入excel

1.增值税发票识别：

import os
import shutil
import sys
import urllib.request
import urllib.parse
import json
import time
import base64

import openpyxl
import xlwt
import pandas  as  pd


# 增值税发票识别/OCR文字识别
def posturl(appcode,file_dir,move_dir,write_dir,write_filename):
    """
    :param file_dir:
    :param write_path:
    :return:
    """
    # file_dir='\\\\'.join(file_dir.split("\\"))+'\\\\'
    # write_path='\\\\'.join(write_path.split("\\"))+'\\\\'
    filenames = []
    err_file = []
    datas_list = []
    datas_title=[]
    for root, dirs, files in os.walk(file_dir):
        for file in files:
            if file.split('.')[-1] in ['jpg', 'png', 'jpeg', 'bmp']:
                filenames.append(file)
    # print(filenames)
    # img_path='C:\\Users\\best\\PycharmProjects\\pythonProject\\timg.jpg'
    for file_name in filenames:
        with open(file_dir + file_name, 'rb') as f:  # 以二进制读取本地图片
            data = f.read()
            encodestr = str(base64.b64encode(data), 'utf-8')
        headers = {
            'Authorization': 'APPCODE %s' % appcode,
            'Content-Type': 'application/json; charset=UTF-8'
        }
        url_request = "https://ocrapi-invoice.taobao.com/ocrservice/invoice"
        dicts = {'img': encodestr}
        try:
            params = json.dumps(dicts).encode(encoding='UTF8')
            req = urllib.request.Request(url_request, params, headers)
            r = urllib.request.urlopen(req)
            html = r.read()
            r.close()
            ret = html.decode("utf8")
            ret_ = json.loads(ret)
            # print(ret_)
            datas_list.append(ret_.get('data'))
            # return ret
        except urllib.error.HTTPError as e:
            err_file.append("%s error" % file_name)
            shutil.move(file_dir + file_name, move_dir)

            # print(e.code)
            # print(e.read().decode("utf8"))
        time.sleep(1)
    fp_title_list = []
    for data_list in datas_list:
        for key in data_list.keys():
            if isinstance(data_list[key], list) and len(data_list[key]) > 0:
                # print(data_list)
                datas_title.append(data_list)
    # 取标题
    datas=datas_title[0]
    for key in datas.keys():
        if isinstance(datas[key], str):
            fp_title_list.append([key])
        elif isinstance(datas[key], list):
            data_ = datas[key][0]
            # print(data_)
            for key_ in data_.keys():
                fp_title_list.append([key_])
        elif isinstance(datas[key], dict):
            # data__ = datas[key]
            for key__ in datas[key]:
                fp_title_list.append([key__])
    for title in fp_title_list:
        for datas in datas_list:
            for key in datas.keys():
                if isinstance(datas[key], str):
                    if key in title:
                        title.append(datas[key])
                elif isinstance(datas[key], list):
                    if len(datas[key]) > 0:
                        data_ = datas[key][0]
                        # print(data_)
                        for key_ in data_.keys():
                            if key_ in title:
                                title.append(data_[key_])
                elif isinstance(datas[key], dict):
                    data__ = datas[key]
                    for key__ in datas[key]:
                        if key__ in title and key__ != '联次':
                            title.append(data__[key__])
    # a = pd.DataFrame(fp_title_list)
    # # print(a)
    # if write_to_excel.lower()=="true" :
    #     a.to_excel(write_dir + write_filename+'.xlsx')

    # write  excel
    f = openpyxl.Workbook()
    # 表名
    f.create_sheet(u'%s' % write_filename)
    sheet1 = f.active
    for index in range(len(fp_title_list)):
        j = 1
        for i in fp_title_list[index]:
            sheet1.cell(j, index + 1, i)
            j += 1
        f.save(write_dir + write_filename + '.xlsx')
    return fp_title_list,str(err_file)


if __name__ == "__main__":
    # 发票目录
    file_dir = 'D:\\fp\\zhfp\\'
    # 写入路径
    write_dir = 'C:\\Users\\best\\Desktop\\'
    # 识别不出来的移动到
    move_dir = 'C:\\Users\\best\\Desktop\\'
    appcode = ' '
    write_filename='new_fp_test.xlsx'
    # appcode=sys.argv[1]     从命令行输入参数的话得用  ""
    # file_dir=sys.argv[2]
    # move_dir=sys.argv[3]
    # write_to_excel=sys.argv[4]
    # write_dir=sys.argv[5]
    # write_filename=sys.argv[6]
    posturl(appcode,file_dir,move_dir,write_dir,write_filename)

2.定额发票识别:

import base64
import json
import os
import time
import urllib, sys
import urllib.request as urllib2
import ssl

import openpyxl
import pandas  as  pd

def Quotalnvoice(appcode,file_dir,write_dir,write_filename):
    filenames = []
    err_file = []
    datas_list = []
    datas_title = []
    fp_title_datas=[]
    host = 'https://ocrapi-quota-invoice.taobao.com'
    path = '/ocrservice/quotaInvoice'
    url = host + path
    method = 'POST'
    for root, dirs, files in os.walk(file_dir):
        for file in files:
            if file.split('.')[-1] in ['jpg', 'png', 'jpeg', 'bmp']:
                filenames.append(file)
    print(filenames)
    for file_name in filenames:
        with open(file_dir + file_name, 'rb') as f:
            data = f.read()
            encodestr = str(base64.b64encode(data), 'utf-8')
        dicts = {'img': encodestr}
        bodys = json.dumps(dicts).encode(encoding='UTF8')
        post_data = bodys
        request = urllib2.Request(url, post_data)
        request.add_header('Authorization', 'APPCODE ' + appcode)
        request.add_header('Content-Type', 'application/json; charset=UTF-8')
        request.add_header('User-Agent','Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36 QIHU 360SE')
        ctx = ssl.create_default_context()
        ctx.check_hostname = False
        ctx.verify_mode = ssl.CERT_NONE
        response = urllib2.urlopen(request, context=ctx)
        content = response.read().decode("utf8")
        time.sleep(1)
        if (content):
            cont = json.loads(content)
            # print(cont)
            result = cont.get('data')
            # print(result)
            datas_list.append(result)
    for data in datas_list:
        for key in data.keys():
            if isinstance(data[key], dict) > 0:
                datas_title.append(data)
    #print(datas_title)
    # take  title
    datas=datas_title[0]
    for key in datas.keys():
        if isinstance(datas[key], str):
            fp_title_datas.append([key])
        elif isinstance(datas[key], dict):
            for key_ in datas[key].keys():
                fp_title_datas.append([key_])
    # print(fp_title_datas)
    # get  data  in  [title,data,...]
    for title in fp_title_datas:
        for datas in datas_list:
            for key in datas.keys():
                if isinstance(datas[key], str):
                    if key in  title:
                        print(type(title))
                        title.append(datas[key])
                elif isinstance(datas[key], dict):
                    for key__ in datas[key].keys():
                        if key__ in  title:
                            title.append(datas[key][key__])
    #pandas write excel
    # a = pd.DataFrame(fp_title_datas)
    # print(a)
    # if write_to_excel.lower()=="true" :
    #     a.to_excel(write_dir + write_filename+'.xlsx')

    # openpyxl  write  excel
    f = openpyxl.Workbook()
    # table_name
    f.create_sheet(u'%s' % write_filename)
    sheet1 = f.active
    for index in range(len(fp_title_datas)):
        j = 1
        for i in fp_title_datas[index]:
            sheet1.cell(j, index + 1, i)
            j += 1
        f.save(write_dir + write_filename + '.xlsx')







    # print(datas_list)


if __name__ == '__main__':
    file_dir = 'D:\\fp\\_quto\\'
    appcode = ' '
    write_dir="D:\\fp\\_quto\\"
    write_filename='QuotaInvoice'
    Quotalnvoice(appcode,file_dir,write_dir,write_filename)

3.混贴发票识别：

import base64
import json
import os
import time
import urllib, sys
import urllib.request as urllib2
import ssl

import openpyxl
import pandas  as  pd

def MixedMultiInvoice(appcode,file_dir,write_dir,write_filename):
    filenames = []
    err_file = []
    datas_list = []
    datas_title = []
    fp_title_datas=[]
    host = 'https://ocrapi-mixed-multi-invoice.taobao.com'
    path = '/ocrservice/mixedMultiInvoice'
    url = host + path
    method = 'POST'
    for root, dirs, files in os.walk(file_dir):
        for file in files:
            if file.split('.')[-1] in ['jpg', 'png', 'jpeg', 'bmp']:
                filenames.append(file)
    print(filenames)
    for file_name in filenames:
        with open(file_dir + file_name, 'rb') as f:
            data = f.read()
            encodestr = str(base64.b64encode(data), 'utf-8')
        dicts = {'img': encodestr}
        bodys = json.dumps(dicts).encode(encoding='UTF8')
        post_data = bodys
        request = urllib2.Request(url, post_data)
        request.add_header('Authorization', 'APPCODE ' + appcode)
        request.add_header('Content-Type', 'application/json; charset=UTF-8')
        request.add_header('User-Agent','Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36 QIHU 360SE')
        ctx = ssl.create_default_context()
        ctx.check_hostname = False
        ctx.verify_mode = ssl.CERT_NONE
        response = urllib2.urlopen(request, context=ctx)
        content = response.read().decode("utf8")
        # time.sleep(1)
        if (content):
            cont = json.loads(content)
            result = cont.get('subMsgs')
            # subMsga{}->result[]包含多种发票data->data{}
            for i in result:
                print((i.get('result')).get('data'))
            # print(result)
            datas_list.append(result)
    # print(datas_list)

if __name__ == '__main__':
    file_dir = 'D:\\fp\\mix\\'
    appcode = ''
    write_dir="D:\\fp\\_quto\\"
    write_filename='QuotaInvoice'
    MixedMultiInvoice(appcode,file_dir,write_dir,write_filename)

4.表格识别：

import sys, os
import base64
import time
import json
import urllib.request as urllib2
import base64


def get_img_base64(img_file):
    with open(img_file, 'rb') as infile:
        s = infile.read()
        return str(base64.b64encode(s),'utf-8')


def predict(url, appcode, img_base64, kv_config, old_format):
    if not old_format:
        param = {}
        param['image'] = img_base64
        if kv_config is not None:
            param['configure'] = kv_config
            # print(param)
        body = json.dumps(param).encode(encoding='UTF8')
    else:
        param = {}
        pic = {}
        pic['dataType'] = 50
        pic['dataValue'] = img_base64
        param['image'] = pic

        if kv_config is not None:
            conf = {}
            conf['dataType'] = 50
            conf['dataValue'] = json.dumps(kv_config).encode(encoding='UTF8')
            param['configure'] = conf

        inputs = {"inputs": [param]}
        body = json.dumps(inputs).encode(encoding='UTF8')

    headers = {'Authorization': 'APPCODE %s' % appcode}
    request = urllib2.Request(url=url, headers=headers, data=body)
    try:
        response = urllib2.urlopen(request, timeout=10)
        return response.code, response.headers, response.read().decode("utf8")
    except urllib2.HTTPError as e:
        return e.code, e.headers, e.read()


def demo(appcode,file_dir):
    filenames = []
    for root, dirs, files in os.walk(file_dir):
        for file in files:
            if file.split('.')[-1] in ['jpg', 'png', 'jpeg', 'bmp']:
                filenames.append(file)
    for filename in filenames:
        url = 'https://form.market.alicloudapi.com/api/predict/ocr_table_parse'
        img_file = file_dir+filename
        # 如果输入带有inputs, 设置为True，否则设为False
        is_old_format = False
        config = {'format': 'xlsx', 'finance': False, 'dir_assure': False}
        # not  config   config=none
        # config = None

        img_base64data = get_img_base64(img_file)
        stat, header, content = predict(url, appcode, img_base64data, config, is_old_format)
        if stat != 200:
            print('Http status code: ', stat)
            print('Error msg in header: ', header['x-ca-error-message'] if 'x-ca-error-message' in header else '')
            print('Error msg in body: ', content)
            exit()
        if is_old_format:
            result_str = json.loads(content)['outputs'][0]['outputValue']['dataValue']
        else:
            result_str = content
        # read  table  content
        result=json.loads(result_str)
        filename=filename.replace('.','_')
        # write excel
        with open(file_dir+'%s.xlsx' %filename, 'wb') as fout:
            # write  content  in output.xlsx
            fout.write(base64.b64decode(result.get("tables")))
        # print(result.get("tables"))
        # result = json.loads(result_str)


if __name__ == '__main__':
    appcode = ' '
    file_dir='D:\\fp\\table\\'
    demo(appcode,file_dir)

用pyinstaller -F .py文件打包成exe，遍历文件夹的时候os.walk(file_dir)，
file_dir 传参必须得"fiel_dir"，‘file_dir’ 函数里都能能读出来，但传不进os.walk()