keithhon commited on
Commit
802b015
·
1 Parent(s): 23ec29f

Upload app.py with huggingface_hub

Browse files
Files changed (1) hide show
  1. app.py +28 -0
app.py ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import os
2
+ import gradio as gr
3
+
4
+ print(os.popen(f'cat /etc/debian_version').read())
5
+ print(os.popen(f'cat /etc/issue').read())
6
+ print(os.popen(f'apt search tesseract').read())
7
+
8
+ choices = os.popen('tesseract --list-langs').read().split('\n')[1:-1]
9
+
10
+ def inference(filepath, languages):
11
+ print('languages', languages)
12
+ languages_str = ' -l ' + '+'.join(languages) if languages else ''
13
+ print('languages_str', languages_str)
14
+ return os.popen(f'tesseract {filepath} -{languages_str}').read()
15
+
16
+ title = "Tesseract OCR"
17
+ description = "Gradio demo for Tesseract. Tesseract is an open source text recognition (OCR) Engine."
18
+ article = "<p style='text-align: center'><a href='https://tesseract-ocr.github.io/' target='_blank'>Tesseract documentation</a> | <a href='https://github.com/tesseract-ocr/tesseract' target='_blank'>Github Repo</a></p>"
19
+ gr.Interface(
20
+ inference,
21
+ [gr.inputs.Image(type="filepath", label="Input"), gr.inputs.CheckboxGroup(choices, type="value", default=['eng'], label='language')],
22
+ 'text',
23
+ title=title,
24
+ description=description,
25
+ article=article,
26
+ examples=[['eurotext.png', ['eng']], ['tesseract_sample.png', ['jpn', 'eng']], ['chi.jpg', ['HanS', 'HanT']]]
27
+ ).launch(enable_queue=True,cache_examples=True)
28
+