Python-百度ocr识别

檀俾九

1. #!/usr/bin/python
   
2. : `4 ^4 l0 C# i) s( M# -*- coding: utf-8 -*-
   
3. 2 p9 o3 E0 c$ n& d0 k+ K! }
   
4. 5 N4 a$ y2 C\" w# {. u  T7 l6 jimport base64
   
5. 9 N\" D/ ?' x) I* |: n5 T4 |from os.path import exists
   
6. 4 b\" P3 Y+ q, r; g\" a6 x% X% ufrom tkinter import Tk
   
7. 9 H! W5 n# c) }  vfrom tkinter.filedialog import askopenfilename
   
8. ! B9 F6 w# Y0 a2 y* T% _  y& Ufrom urllib.parse import urlencode
   
9. # P4 \\" Z/ O1 |6 w0 n, @
   
10. , P/ [+ j* X0 u$ R! O- Y1 Timport requests
    
11. . P* h: E+ ?, ?0 p; K9 A9 o0 r7 N2 G; e# i9 j5 Q; b+ q- ^
    
12. 
    
13. 4 R0 }& s' N  f2 Edef Dialog_box():% A3 B- `& T$ V
    
14.     root = Tk()3 a3 T3 h, Q( K2 I
    
15.     root.withdraw()7 u! F2 v3 N+ F  J/ `( O  T; d
    
16. 
    
17. 2 J5 M$ ~4 F0 @4 X& J( _2 s8 O6 k    filetypes = [(+ S* e5 [+ K7 F& s$ K' f% @: H
    
18.         '图片文件(*.jpg;*.jpeg;*.gif;*.png;*.bmp)', '*.jpg;*.jpeg;*.gif;*.png;*.bmp'),
    
19. 5 P* q1 R: y# V. L4 _3 b. p6 ?        ('所有文件(*.*)', '*.*')]
    
20. + q3 Y( a9 X0 u* f( K: v( g3 n1 b: X
    
21.     file_path = askopenfilename(filetypes=filetypes)$ X/ ^0 P2 c6 @) t; C
    
22.     return file_path0 v7 H, |0 p9 R# l$ `
    
23. ( g: Q+ ?$ ~3 `1 _& o! r5 Q7 [/ j
    
24. 8 o9 q\" l& ^  w3 r; }
    
25. def Read_file(file_path):
    
26. 6 c! V- T  t7 ^' D( [  {' x    with open(file_path, mode='rb') as file_object:$ r. H; O* h3 F7 K
    
27.         contents = file_object.read(): |0 g; k- D: X8 ?6 \2 ]
    
28.         return contents.rstrip()
    
29. : t) I8 t7 y; t$ g& d! n- X+ J, o3 _/ J4 {# \$ D, j; I5 L
    
30. / r# f& X  y1 Y5 b, `! g1 V
    
31. def Baidu_ocr(binary):
    
32. & u! G2 z  O9 U. C6 z: |$ @3 Q    url = "https://aip.baidubce.com/oauth/2.0/token?": v* J2 t3 K. a6 m
    
33.     data = {
    
34. 1 f! E. K! a5 t) x        "grant_type": "client_credentials",
    
35. - E5 Q6 L; ^6 K0 d/ A        "client_id": "",  # 必填# {* a. U\" u0 e0 t3 a
    
36.         "client_secret": ""  # 必填
    
37. , \) f# N5 F  h/ q' m    }8 q5 R' w) s+ N1 _1 k
    
38.     url += urlencode(data)
    
39. 8 {, r) B! _& r* }- L3 O1 A    response = requests.get(url=url).json()
    
40. : v, H; u7 C% B  C5 `\" r    access_token = response['access_token']
    
41. , g9 K9 Z- C0 g; e
    
42. ' k+ L; ^\" I0 l% O- t- ~  x& Z, |* ?    url = "https://aip.baidubce.com/rest/2.0/ocr/v1/general_basic?access_token=" + access_token
    
43. ) d0 W7 t; K1 |0 G. `    post_data = {
    
44. ; B2 F  R- p3 f, x. h( a        "image": base64.b64encode(binary),
    
45. 7 c9 D3 q* D; y( g- @% [5 A0 q        "url": "",
    
46.   p( M! Q' u! O2 Z        "language_type": "CHN_ENG",3 r; q7 k& S! Y% V/ H1 B0 w
    
47.         "detect_direction": "false",& b& Z, p\" V& i; `( t
    
48.         "detect_language": "false",; H; T9 I* U6 B: W. R
    
49.         "probability": "false"; y) k# X5 @; E! r+ v8 Z1 u/ B
    
50.     }
    
51. & r% F* S8 w: ^' S    response = requests.post(url=url, data=post_data).json()
    
52. / `( J: C% G: k! T4 k    num = response['words_result_num']  L* E; t( x7 U
    
53.     result = ''
    
54. \" s4 W* n: S* q) D\" l/ a& c    for i in range(num):& s# d7 u9 h* [  Z\" v0 [) x
    
55.         result += response['words_result'][ i]['words'] + "\r\n"
    
56.   A2 L+ [) E) S3 X0 V$ H    return result
    
57. \" f9 V* D! R, C
    
58. + @9 @9 @. j4 B, n
    
59. \" B\" B7 y/ W1 L% ?! n/ F5 {# @if __name__ == "__main__":
    
60. 9 U! L( F  `' n1 Z5 e0 G& o# A    file_path = Dialog_box()+ j: r: t4 I0 f4 F
    
61.     if exists(file_path):0 Y/ }2 J# H. {( D5 [
    
62.         data = Read_file(file_path)
    
63. 5 M; u6 ]& ^. Z/ P        print(Baidu_ocr(data))
    
64. 8 O% B% Y, d$ z1 y) P5 c$ F0 |) X, L, ~/ L( P' t
    

#!/usr/bin/python<br /> <span style=\"display:none\">: `4 ^4 l0 C# i) s( M</span># -*- coding: utf-8 -*-<br /> <span style=\"display:none\">2 p9 o3 E0 c$ n& d0 k+ K! }</span><br /> <span style=\"display:none\">5 N4 a$ y2 C\" w# {. u  T7 l6 j</span>import base64<br /> <span style=\"display:none\">9 N\" D/ ?' x) I* |: n5 T4 |</span>from os.path import exists<br /> <span style=\"display:none\">4 b\" P3 Y+ q, r; g\" a6 x% X% u</span>from tkinter import Tk<br /> <span style=\"display:none\">9 H! W5 n# c) }  v</span>from tkinter.filedialog import askopenfilename<br /> <span style=\"display:none\">! B9 F6 w# Y0 a2 y* T% _  y& U</span>from urllib.parse import urlencode<br /> <span style=\"display:none\"># P4 \\" Z/ O1 |6 w0 n, @</span><br /> <span style=\"display:none\">, P/ [+ j* X0 u$ R! O- Y1 T</span>import requests<br /> <span style=\"display:none\">. P* h: E+ ?, ?0 p</span><font class=\"jammer\">; K9 A9 o0 r7 N2 G; e# i9 j5 Q; b+ q- ^</font><br /> <br /> <span style=\"display:none\">4 R0 }& s' N  f2 E</span>def Dialog_box():<font class=\"jammer\">% A3 B- `& T$ V</font><br />     root = Tk()<font class=\"jammer\">3 a3 T3 h, Q( K2 I</font><br />     root.withdraw()<font class=\"jammer\">7 u! F2 v3 N+ F  J/ `( O  T; d</font><br /> <br /> <span style=\"display:none\">2 J5 M$ ~4 F0 @4 X& J( _2 s8 O6 k</span>    filetypes = [(<font class=\"jammer\">+ S* e5 [+ K7 F& s$ K' f% @: H</font><br />         '图片文件(*.jpg;*.jpeg;*.gif;*.png;*.bmp)', '*.jpg;*.jpeg;*.gif;*.png;*.bmp'),<br /> <span style=\"display:none\">5 P* q1 R: y# V. L4 _3 b. p6 ?</span>        ('所有文件(*.*)', '*.*')]<br /> <span style=\"display:none\">+ q3 Y( a9 X0 u* f</span><font class=\"jammer\">( K: v( g3 n1 b: X</font><br />     file_path = askopenfilename(filetypes=filetypes)<font class=\"jammer\">$ X/ ^0 P2 c6 @) t; C</font><br />     return file_path<font class=\"jammer\">0 v7 H, |0 p9 R# l$ `</font><br /> <font class=\"jammer\">( g: Q+ ?$ ~3 `1 _& o! r5 Q7 [/ j</font><br /> <font class=\"jammer\">8 o9 q\" l& ^  w3 r; }</font><br /> def Read_file(file_path):<br /> <span style=\"display:none\">6 c! V- T  t7 ^' D( [  {' x</span>    with open(file_path, mode='rb') as file_object:<font class=\"jammer\">$ r. H; O* h3 F7 K</font><br />         contents = file_object.read()<font class=\"jammer\">: |0 g; k- D: X8 ?6 \2 ]</font><br />         return contents.rstrip()<br /> <span style=\"display:none\">: t) I8 t7 y; t$ g& d! n</span><font class=\"jammer\">- X+ J, o3 _/ J4 {# \$ D, j; I5 L</font><br /> <font class=\"jammer\">/ r# f& X  y1 Y5 b, `! g1 V</font><br /> def Baidu_ocr(binary):<br /> <span style=\"display:none\">& u! G2 z  O9 U. C6 z: |$ @3 Q</span>    url = "https://aip.baidubce.com/oauth/2.0/token?"<font class=\"jammer\">: v* J2 t3 K. a6 m</font><br />     data = {<br /> <span style=\"display:none\">1 f! E. K! a5 t) x</span>        "grant_type": "client_credentials",<br /> <span style=\"display:none\">- E5 Q6 L; ^6 K0 d/ A</span>        "client_id": "",  # 必填<font class=\"jammer\"># {* a. U\" u0 e0 t3 a</font><br />         "client_secret": ""  # 必填<br /> <span style=\"display:none\">, \) f# N5 F  h/ q' m</span>    }<font class=\"jammer\">8 q5 R' w) s+ N1 _1 k</font><br />     url += urlencode(data)<br /> <span style=\"display:none\">8 {, r) B! _& r* }- L3 O1 A</span>    response = requests.get(url=url).json()<br /> <span style=\"display:none\">: v, H; u7 C% B  C5 `\" r</span>    access_token = response['access_token']<br /> <span style=\"display:none\">, g9 K9 Z- C0 g; e</span><br /> <span style=\"display:none\">' k+ L; ^\" I0 l% O- t- ~  x& Z, |* ?</span>    url = "https://aip.baidubce.com/rest/2.0/ocr/v1/general_basic?access_token=" + access_token<br /> <span style=\"display:none\">) d0 W7 t; K1 |0 G. `</span>    post_data = {<br /> <span style=\"display:none\">; B2 F  R- p3 f, x. h( a</span>        "image": base64.b64encode(binary),<br /> <span style=\"display:none\">7 c9 D3 q* D; y( g- @% [5 A0 q</span>        "url": "",<br /> <span style=\"display:none\">  p( M! Q' u! O2 Z</span>        "language_type": "CHN_ENG",<font class=\"jammer\">3 r; q7 k& S! Y% V/ H1 B0 w</font><br />         "detect_direction": "false",<font class=\"jammer\">& b& Z, p\" V& i; `( t</font><br />         "detect_language": "false",<font class=\"jammer\">; H; T9 I* U6 B: W. R</font><br />         "probability": "false"<font class=\"jammer\">; y) k# X5 @; E! r+ v8 Z1 u/ B</font><br />     }<br /> <span style=\"display:none\">& r% F* S8 w: ^' S</span>    response = requests.post(url=url, data=post_data).json()<br /> <span style=\"display:none\">/ `( J: C% G: k! T4 k</span>    num = response['words_result_num']<font class=\"jammer\">  L* E; t( x7 U</font><br />     result = ''<br /> <span style=\"display:none\">\" s4 W* n: S* q) D\" l/ a& c</span>    for i in range(num):<font class=\"jammer\">& s# d7 u9 h* [  Z\" v0 [) x</font><br />         result += response['words_result'][ i]['words'] + "\r\n"<br /> <span style=\"display:none\">  A2 L+ [) E) S3 X0 V$ H</span>    return result<br /> <span style=\"display:none\">\" f9 V* D! R, C</span><br /> <span style=\"display:none\">+ @9 @9 @. j4 B, n</span><br /> <span style=\"display:none\">\" B\" B7 y/ W1 L% ?! n/ F5 {# @</span>if __name__ == "__main__":<br /> <span style=\"display:none\">9 U! L( F  `' n1 Z5 e0 G& o# A</span>    file_path = Dialog_box()<font class=\"jammer\">+ j: r: t4 I0 f4 F</font><br />     if exists(file_path):<font class=\"jammer\">0 Y/ }2 J# H. {( D5 [</font><br />         data = Read_file(file_path)<br /> <span style=\"display:none\">5 M; u6 ]& ^. Z/ P</span>        print(Baidu_ocr(data))<br /> <span style=\"display:none\">8 O% B% Y, d$ z1 y) P5 c</span><font class=\"jammer\">$ F0 |) X, L, ~/ L( P' t</font><br />

$ M1 N1 B; M: g) e7 n
3 @" F* A* m$ ]  D8 u4 N* A/ W" i0 Q

转发自派生社区

Python交流群：1047602540