xzbautomarkapi/utils/bdocr.py

#-*-coding:utf-8 -*-

import json,re
from aip import AipOcr

APP_ID = '11225532'
API_KEY = 'iGhZP96p570OG62oHPRExG09'
SECRET_KEY = 'XGTCHsTf7sWkVmBNKNoGIcWfGXrGHYHI'

client = AipOcr(APP_ID, API_KEY, SECRET_KEY)

def get_file_content(filePath):
	with open(filePath, 'rb') as fp:
		return fp.read()

def rec_basicgeneral(imgpath):
    """
    通用文字识别
    """
    words = []
    image = get_file_content(imgpath)
    #res_image = client.basicGeneral(image)
    res_image = client.general(image)
    for item in res_image["words_result"]:
        print(item)
        if item["words"]:
            words.append(item)
    return words

def rec_general_num(imgpath):
    """
    识别出数字(题号)
    """
    nums = []
    nums_dct = {}
    choices = set()
    words = rec_basicgeneral(imgpath)
    for word in words:
        location = word["location"]
        word = word["words"]
        res = re.match(r"^(\d+).*",word)
        if res:
            nums.append(int(res.group(1)))
            nums_dct[int(res.group(1))] = location
        res = re.findall(r"(A|B|C|D|E|F|G)",word)
        if res:
            choices.update(set(res))
    print(nums)
    if (1 in nums) and (2 in nums) and (nums.index(2)-nums.index(1) > 1):
        rank_order = 1
    else:
        rank_order = 2
    nums.sort()
    return nums,nums_dct,list(choices),rank_order

if __name__ == "__main__":
    imgpath = "/tmp/src4.png"
    imgpath = "/tmp/test_0_crop.png"
    nums,nums_dct,choices,rank_order = rec_general_num(imgpath)
    print(nums,nums_dct,choices,rank_order,222222222222222222)