Python-百度ocr识别

檀俾九

1 N3 c# ]* H& d$ K2 v! P

1. #!/usr/bin/python
   
2. \" F+ w/ m* u+ D2 ^- z% K# -*- coding: utf-8 -*-: r5 f: S5 ~, h
   
3. 
   
4. ) Q$ r$ r& l/ i' z& N7 c5 C5 fimport base64# X  m0 D4 ]4 P7 N1 A( o
   
5. from os.path import exists
   
6. 8 x: V9 b/ C5 S% @: Z4 Afrom tkinter import Tk9 y% f' D. `- O  N4 a
   
7. from tkinter.filedialog import askopenfilename
   
8. 5 u# {* j6 x! c' E: X7 Ffrom urllib.parse import urlencode* V* D4 f3 L# G! n+ D7 h+ A
   
9. ; b3 d. B: X\" @9 p  f! h. g
   
10. import requests6 h, j: G! z4 d- s) ^
    
11. 
    
12. 1 G  F/ q! m  Z$ _* n8 A5 ]8 Y
    
13. - {# P' {# R, V( S. H2 S7 Ldef Dialog_box():; A. I4 o8 F+ E. N, l. Q9 \
    
14.     root = Tk()
    
15. ; `0 H0 y: N# i+ `5 D8 w* w+ @    root.withdraw()$ x8 Z1 _9 a) V$ b
    
16. ' x+ F6 }9 o& n8 U3 Y
    
17.     filetypes = [(
    
18. * n# ]+ p\" u; ^        '图片文件(*.jpg;*.jpeg;*.gif;*.png;*.bmp)', '*.jpg;*.jpeg;*.gif;*.png;*.bmp'),
    
19. ' h# c' u+ C. j% x  u        ('所有文件(*.*)', '*.*')]
    
20. . h& i8 Z1 z( O- K, t* s2 ]/ y+ N/ J' @/ J
    
21.     file_path = askopenfilename(filetypes=filetypes), h\" }/ H; q7 e+ }- Q% z
    
22.     return file_path
    
23. 6 r5 R* ]6 X6 W# C+ ?
    
24. 5 A( R( }0 M0 h5 `' F4 K  `$ @7 B8 Y7 f+ h8 A6 P1 [
    
25. def Read_file(file_path):4 r6 Q' p# i: a2 a% O
    
26.     with open(file_path, mode='rb') as file_object:
    
27. , H/ L/ `& i  {! R4 K        contents = file_object.read()1 I3 B' A' s: h4 }( ~
    
28.         return contents.rstrip()\" D) F- x/ P* w* R5 j/ W1 m
    
29. . B4 K7 G) k( V
    
30. 
    
31. ) o) t. a' `0 _def Baidu_ocr(binary):
    
32. , W3 I4 t  U1 f8 a6 a% w% }0 N    url = "https://aip.baidubce.com/oauth/2.0/token?"
    
33. - @) T% c( f( p% s% [    data = {
    
34. * v! p\" u/ ?  i3 {0 U        "grant_type": "client_credentials",# k& e1 f0 ^' d6 ^9 S5 P) [- X
    
35.         "client_id": "",  # 必填
    
36. ! r/ ~/ a. q. k9 F6 F        "client_secret": ""  # 必填
    
37. % C' Q& W6 P5 `) K* X- B    }
    
38. * t! A! L2 A3 F0 f' _    url += urlencode(data)) U& p9 \0 N1 k, {- b5 z
    
39.     response = requests.get(url=url).json(): ~, g1 V% l3 A* G* g7 ]3 L6 j9 Q% Y' t
    
40.     access_token = response['access_token']
    
41. 1 J\" ~5 K# z# p  x) }# c8 ~( |9 |3 T5 `/ H: ^  q8 W0 r( H
    
42.     url = "https://aip.baidubce.com/rest/2.0/ocr/v1/general_basic?access_token=" + access_token
    
43. 5 n# Y0 L5 W3 t! M: B\" ?    post_data = {
    
44. 1 u# m( u, {; ]) K        "image": base64.b64encode(binary),2 A: m* V4 u7 I! A
    
45.         "url": "",% ^5 y6 q  J$ R. D% g8 \3 M
    
46.         "language_type": "CHN_ENG",8 z% ~5 r% `; z
    
47.         "detect_direction": "false",
    
48. * u2 l' m\" h/ D, x: x, J        "detect_language": "false",
    
49. \" V+ \* d7 ^4 g( t        "probability": "false"
    
50. 1 b+ @) D9 n8 U/ \5 k4 S* C    }6 ^: @2 c4 ?6 s9 v! n
    
51.     response = requests.post(url=url, data=post_data).json()
    
52. / F  v, X4 A! ]' a1 t; u    num = response['words_result_num']- z8 y6 B9 }* t7 O8 H) }1 u' p! Z
    
53.     result = ''
    
54. 4 W# }6 Y0 L& [( j    for i in range(num):' K  A  O: ?\" J$ i9 s( Z* v0 `
    
55.         result += response['words_result'][ i]['words'] + "\r\n"
    
56. 2 f0 q& e. A3 }! `( b0 [  @    return result
    
57. 8 Q! n0 `5 l! ?! C- C. ]3 @5 V( c3 h+ s6 t3 E3 X
    
58. # z* e3 S% H; Z\" n; o; M6 {
    
59. if __name__ == "__main__":
    
60. % M7 Y/ c! Q7 L% {! F. T    file_path = Dialog_box()
    
61. \" [8 ~' d( i' R6 P1 M    if exists(file_path):
    
62. 1 m- C9 t; ^7 @9 ?        data = Read_file(file_path)+ X' J/ I7 K, X) _4 V. A
    
63.         print(Baidu_ocr(data))
    
64. 7 e, _' f5 ?' J( C4 d: W& ~. R5 M; V6 i0 T# {
    

#!/usr/bin/python<br /> <span style=\"display:none\">\" F+ w/ m* u+ D2 ^- z% K</span># -*- coding: utf-8 -*-<font class=\"jammer\">: r5 f: S5 ~, h</font><br /> <br /> <span style=\"display:none\">) Q$ r$ r& l/ i' z& N7 c5 C5 f</span>import base64<font class=\"jammer\"># X  m0 D4 ]4 P7 N1 A( o</font><br /> from os.path import exists<br /> <span style=\"display:none\">8 x: V9 b/ C5 S% @: Z4 A</span>from tkinter import Tk<font class=\"jammer\">9 y% f' D. `- O  N4 a</font><br /> from tkinter.filedialog import askopenfilename<br /> <span style=\"display:none\">5 u# {* j6 x! c' E: X7 F</span>from urllib.parse import urlencode<font class=\"jammer\">* V* D4 f3 L# G! n+ D7 h+ A</font><br /> <font class=\"jammer\">; b3 d. B: X\" @9 p  f! h. g</font><br /> import requests<font class=\"jammer\">6 h, j: G! z4 d- s) ^</font><br /> <br /> <span style=\"display:none\">1 G  F/ q! m  Z$ _* n8 A5 ]8 Y</span><br /> <span style=\"display:none\">- {# P' {# R, V( S. H2 S7 L</span>def Dialog_box():<font class=\"jammer\">; A. I4 o8 F+ E. N, l. Q9 \</font><br />     root = Tk()<br /> <span style=\"display:none\">; `0 H0 y: N# i+ `5 D8 w* w+ @</span>    root.withdraw()<font class=\"jammer\">$ x8 Z1 _9 a) V$ b</font><br /> <font class=\"jammer\">' x+ F6 }9 o& n8 U3 Y</font><br />     filetypes = [(<br /> <span style=\"display:none\">* n# ]+ p\" u; ^</span>        '图片文件(*.jpg;*.jpeg;*.gif;*.png;*.bmp)', '*.jpg;*.jpeg;*.gif;*.png;*.bmp'),<br /> <span style=\"display:none\">' h# c' u+ C. j% x  u</span>        ('所有文件(*.*)', '*.*')]<br /> <span style=\"display:none\">. h& i8 Z1 z( O</span><font class=\"jammer\">- K, t* s2 ]/ y+ N/ J' @/ J</font><br />     file_path = askopenfilename(filetypes=filetypes)<font class=\"jammer\">, h\" }/ H; q7 e+ }- Q% z</font><br />     return file_path<br /> <span style=\"display:none\">6 r5 R* ]6 X6 W# C+ ?</span><br /> <span style=\"display:none\">5 A( R( }0 M0 h5 `' F4 K  `</span><font class=\"jammer\">$ @7 B8 Y7 f+ h8 A6 P1 [</font><br /> def Read_file(file_path):<font class=\"jammer\">4 r6 Q' p# i: a2 a% O</font><br />     with open(file_path, mode='rb') as file_object:<br /> <span style=\"display:none\">, H/ L/ `& i  {! R4 K</span>        contents = file_object.read()<font class=\"jammer\">1 I3 B' A' s: h4 }( ~</font><br />         return contents.rstrip()<font class=\"jammer\">\" D) F- x/ P* w* R5 j/ W1 m</font><br /> <font class=\"jammer\">. B4 K7 G) k( V</font><br /> <br /> <span style=\"display:none\">) o) t. a' `0 _</span>def Baidu_ocr(binary):<br /> <span style=\"display:none\">, W3 I4 t  U1 f8 a6 a% w% }0 N</span>    url = "https://aip.baidubce.com/oauth/2.0/token?"<br /> <span style=\"display:none\">- @) T% c( f( p% s% [</span>    data = {<br /> <span style=\"display:none\">* v! p\" u/ ?  i3 {0 U</span>        "grant_type": "client_credentials",<font class=\"jammer\"># k& e1 f0 ^' d6 ^9 S5 P) [- X</font><br />         "client_id": "",  # 必填<br /> <span style=\"display:none\">! r/ ~/ a. q. k9 F6 F</span>        "client_secret": ""  # 必填<br /> <span style=\"display:none\">% C' Q& W6 P5 `) K* X- B</span>    }<br /> <span style=\"display:none\">* t! A! L2 A3 F0 f' _</span>    url += urlencode(data)<font class=\"jammer\">) U& p9 \0 N1 k, {- b5 z</font><br />     response = requests.get(url=url).json()<font class=\"jammer\">: ~, g1 V% l3 A* G* g7 ]3 L6 j9 Q% Y' t</font><br />     access_token = response['access_token']<br /> <span style=\"display:none\">1 J\" ~5 K# z# p  x) }# c8 ~( |</span><font class=\"jammer\">9 |3 T5 `/ H: ^  q8 W0 r( H</font><br />     url = "https://aip.baidubce.com/rest/2.0/ocr/v1/general_basic?access_token=" + access_token<br /> <span style=\"display:none\">5 n# Y0 L5 W3 t! M: B\" ?</span>    post_data = {<br /> <span style=\"display:none\">1 u# m( u, {; ]) K</span>        "image": base64.b64encode(binary),<font class=\"jammer\">2 A: m* V4 u7 I! A</font><br />         "url": "",<font class=\"jammer\">% ^5 y6 q  J$ R. D% g8 \3 M</font><br />         "language_type": "CHN_ENG",<font class=\"jammer\">8 z% ~5 r% `; z</font><br />         "detect_direction": "false",<br /> <span style=\"display:none\">* u2 l' m\" h/ D, x: x, J</span>        "detect_language": "false",<br /> <span style=\"display:none\">\" V+ \* d7 ^4 g( t</span>        "probability": "false"<br /> <span style=\"display:none\">1 b+ @) D9 n8 U/ \5 k4 S* C</span>    }<font class=\"jammer\">6 ^: @2 c4 ?6 s9 v! n</font><br />     response = requests.post(url=url, data=post_data).json()<br /> <span style=\"display:none\">/ F  v, X4 A! ]' a1 t; u</span>    num = response['words_result_num']<font class=\"jammer\">- z8 y6 B9 }* t7 O8 H) }1 u' p! Z</font><br />     result = ''<br /> <span style=\"display:none\">4 W# }6 Y0 L& [( j</span>    for i in range(num):<font class=\"jammer\">' K  A  O: ?\" J$ i9 s( Z* v0 `</font><br />         result += response['words_result'][ i]['words'] + "\r\n"<br /> <span style=\"display:none\">2 f0 q& e. A3 }! `( b0 [  @</span>    return result<br /> <span style=\"display:none\">8 Q! n0 `5 l! ?! C- C</span><font class=\"jammer\">. ]3 @5 V( c3 h+ s6 t3 E3 X</font><br /> <font class=\"jammer\"># z* e3 S% H; Z\" n; o; M6 {</font><br /> if __name__ == "__main__":<br /> <span style=\"display:none\">% M7 Y/ c! Q7 L% {! F. T</span>    file_path = Dialog_box()<br /> <span style=\"display:none\">\" [8 ~' d( i' R6 P1 M</span>    if exists(file_path):<br /> <span style=\"display:none\">1 m- C9 t; ^7 @9 ?</span>        data = Read_file(file_path)<font class=\"jammer\">+ X' J/ I7 K, X) _4 V. A</font><br />         print(Baidu_ocr(data))<br /> <span style=\"display:none\">7 e, _' f5 ?' J( C4 d</span><font class=\"jammer\">: W& ~. R5 M; V6 i0 T# {</font><br />

( E* \2 J) K7 f! n. i

转发自派生社区

Python交流群：1047602540