1# coding=utf-8
2
3import sys
4import json
5import base64
6
7
8# 保证兼容python2以及python3
9IS_PY3 = sys.version_info.major == 3
10if IS_PY3:
11    from urllib.request import urlopen
12    from urllib.request import Request
13    from urllib.error import URLError
14    from urllib.parse import urlencode
15    from urllib.parse import quote_plus
16else:
17    import urllib2
18    from urllib import quote_plus
19    from urllib2 import urlopen
20    from urllib2 import Request
21    from urllib2 import URLError
22    from urllib import urlencode
23
24# 防止https证书校验不正确
25import ssl
26ssl._create_default_https_context = ssl._create_unverified_context
27
28API_KEY = 'GmhC18eVP1Fo1ECX911dtOzw'
29
30SECRET_KEY = 'PQ2ukO4Aec2PTsgQU9UkiEKYciavlZk8'
31
32
33OCR_URL = "https://aip.baidubce.com/rest/2.0/ocr/v1/accurate_basic"
34
35
36"""  TOKEN start """
37TOKEN_URL = 'https://aip.baidubce.com/oauth/2.0/token'
38
39
40"""
41    获取token
42"""
43def fetch_token():
44    params = {'grant_type': 'client_credentials',
45              'client_id': API_KEY,
46              'client_secret': SECRET_KEY}
47    post_data = urlencode(params)
48    if (IS_PY3):
49        post_data = post_data.encode('utf-8')
50    req = Request(TOKEN_URL, post_data)
51    try:
52        f = urlopen(req, timeout=5)
53        result_str = f.read()
54    except URLError as err:
55        print(err)
56    if (IS_PY3):
57        result_str = result_str.decode()
58
59
60    result = json.loads(result_str)
61
62    if ('access_token' in result.keys() and 'scope' in result.keys()):
63        if not 'brain_all_scope' in result['scope'].split(' '):
64            print ('please ensure has check the  ability')
65            exit()
66        return result['access_token']
67    else:
68        print ('please overwrite the correct API_KEY and SECRET_KEY')
69        exit()
70
71"""
72    读取文件
73"""
74def read_file(image_path):
75    f = None
76    try:
77        f = open(image_path, 'rb')
78        return f.read()
79    except:
80        print('read image file fail')
81        return None
82    finally:
83        if f:
84            f.close()
85
86
87"""
88    调用远程服务
89"""
90def request(url, data):
91    req = Request(url, data.encode('utf-8'))
92    has_error = False
93    try:
94        f = urlopen(req)
95        result_str = f.read()
96        if (IS_PY3):
97            result_str = result_str.decode()
98        return result_str
99    except  URLError as err:
100        print(err)
101
102if __name__ == '__main__':
103
104    # 获取access token
105    token = fetch_token()
106
107    # 拼接通用文字识别高精度url
108    image_url = OCR_URL + "?access_token=" + token
109
110    text = ""
111
112    # 读取测试图片
113    file_content = read_file('./text.jpg')
114
115    # 调用文字识别服务
116    result = request(image_url, urlencode({'image': base64.b64encode(file_content)}))
117
118    # 解析返回结果
119    result_json = json.loads(result)
120    for words_result in result_json["words_result"]:
121        text = text + words_result["words"]
122
123    # 打印文字
124    print(text)