1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
|
import io
import os
import sys
import requests
import base64
import time
import tkinter as tk
from tkinter import filedialog
from PIL import ImageGrab
requests.adapters.DEFAULT_RETRIES = 2
basic_url = 'https://aip.baidubce.com/'
def get_token():
host = basic_url+'oauth/2.0/token?grant_type=client_credentials&client_id=xxx&client_secret=xxx'
try:
with open(sys.path[0]+"/token") as t:
return t.read()
except:
print("token不存在！准备获取..")
try:
response = requests.get(host)
aTok = response.json()['access_token']
with open(sys.path[0]+"/token", "w") as code:
code.write(aTok)
except:
input('无法获取或保存token！Ctrl+C取消重试')
return get_token()
else:
print('获取token完成！')
return aTok
def send_img():
'''
表格文字识别(异步接口)
'''
request_url = basic_url+"rest/2.0/solution/v1/form_ocr/request"
request_url = request_url + "?access_token=" + token
headers = {'content-type': 'application/x-www-form-urlencoded'}
print('发送图片中..')
try:
response = requests.post(request_url, data=imgdata, headers=headers, timeout=(5, TOUT))
print(response.text)
requestId = response.json()['result'][0]['request_id']
print('图片上传完成！')
return(requestId)
except:
print(sys.exc_info())
input('上传失败，请尝试直接从网页复制图片！ Ctrl+C取消重试')
return send_img()
def fetch_result(request_id):
'''
表格文字识别(异步接口)
'''
request_url = basic_url+"rest/2.0/solution/v1/form_ocr/get_request_result"
params = {"request_id": request_id,'result_type':'excel'}
request_url = request_url + "?access_token=" + token
headers = {'content-type': 'application/x-www-form-urlencoded'}
counts = 1
tags=['—','\\','|','/']
while (counts<=30):
try:
counts+=1
response = requests.post(request_url, data=params, headers=headers)
percent = int(response.json()['result']['percent'])
sys.stdout.write("\r识别中[%s]%3d%%|%s| %s/100" %(tags[(percent-1)%4], percent, "█"*(percent//2), percent))
sys.stdout.flush()
if percent == 100:
break
except:
pass
time.sleep(2)
print()
try:
result_data=response.json()['result']['result_data']
except:
input('无法获取识别状态！Ctrl+C取消重试')
return fetch_result(request_id)
else:
print('识别完成->\033[4m{0}\033[0m'.format(result_data))
return(result_data)
def download_xls(url):
'''
表格文字识别(异步接口)
'''
print('开始下载xls..')
try:
r = requests.get(url)
root = tk.Tk()
root.withdraw()
with filedialog.asksaveasfile(mode='wb', defaultextension=".xls", filetypes=[('Excel','.xls')]) as f:
if f:
f.write(r.content)
print('表格下载完成！')
except:
input('下载失败！Ctrl+C取消重试')
return download_xls(url)
return 0
def accurate_basic():
'''
文字精确识别(同步接口)
'''
request_url = basic_url+"rest/2.0/ocr/v1/accurate_basic"
access_token = token
request_url = request_url + "?access_token=" + access_token
headers = {'content-type': 'application/x-www-form-urlencoded'}
try:
response = requests.post(request_url, data=imgdata, headers=headers)
print('识别完成！\n==========')
retString = ''
for words in response.json()['words_result']:
retString=retString+words['words']+'\n'
return retString
except:
input('识别错误！Ctrl+C取消重试')
return accurate_basic()
def parse_img():
global imgdata
str2 = "表格" if ISTBL else "文本"
if IMG == '':
try:
clipimg = ImageGrab.grabclipboard()
img_bytes = io.BytesIO()
clipimg.save(img_bytes, format='PNG')
imgdata = {"image": base64.b64encode(img_bytes.getvalue())}
return '剪切板', str2
except:
input('从剪切板获取失败！Ctrl+C取消重试')
return parse_img()
elif IMG == 'f':
root = tk.Tk()
root.withdraw()
try:
with filedialog.askopenfile(mode="rb", filetypes=[('image files',('.png','.jpg','.jpeg','.bmp'))]) as f:
if f:
imgdata = {"image": base64.b64encode(f.read())}
return f.name, str2
except:
input('选择图片文件失败！Ctrl+C取消重试')
return parse_img()
elif IMG.startswith('http'):
if ISTBL:
try:
resp = requests.get(IMG)
imgdata = {"image": base64.b64encode(resp.content)}
except:
input('图片下载失败！Ctrl+C取消重试')
return parse_img()
else:
imgdata = {"url": IMG}
return '\033[4m{0}\033[0m'.format(IMG), str2
TOUT=20
imgdata = {}
os.system("color")
token=get_token()
while 1:
IMG=input('输入链接(留空=剪切板；f=文件选择；q=退出)：')
if IMG == 'q':
break
ISTBL=input("表格请输入文字再回车：")
imgsrc, imgtype = parse_img()
input('从 \033[4m{0}\033[0m 识别 \033[4m{1}\033[0m，回车确认上传..'.format(imgsrc, imgtype))
if ISTBL:
request_id=send_img()
url=fetch_result(request_id)
download_xls(url)
else:
print(accurate_basic())
|
（貌似百度不咋受开源社区待见)。nodejs几乎不存在，python的倒是找到一些简单脚本。以下记录找到的有用信息：
，没有轮询，放弃
