Spaces:
Build error
Build error
amoldwalunj
commited on
Commit
•
7902e1a
1
Parent(s):
38ad06f
Create app.py
Browse files
app.py
ADDED
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import pytesseract
|
2 |
+
from PIL import Image
|
3 |
+
import pandas as pd
|
4 |
+
import os
|
5 |
+
import gradio as gr
|
6 |
+
import numpy as np
|
7 |
+
import gradio as gr
|
8 |
+
|
9 |
+
|
10 |
+
def ocr_df_using_pytesseract(image):
|
11 |
+
pytesseract.pytesseract.tesseract_cmd =r"C:\Users\amold\Desktop\Upwork\pdf to image and pytesseract\tesseact_exe\Tesseract-OCR\tesseract.exe"
|
12 |
+
|
13 |
+
#image = Image.open(example['image_path'])
|
14 |
+
|
15 |
+
width, height = image.size
|
16 |
+
|
17 |
+
# apply ocr to the image
|
18 |
+
ocr_df = pytesseract.image_to_data(image, output_type='data.frame')
|
19 |
+
float_cols = ocr_df.select_dtypes('float').columns
|
20 |
+
ocr_df = ocr_df.dropna().reset_index(drop=True)
|
21 |
+
ocr_df[float_cols] = ocr_df[float_cols].round(0).astype(int)
|
22 |
+
ocr_df = ocr_df.replace(r'^\s*$', np.nan, regex=True)
|
23 |
+
ocr_df = ocr_df.dropna().reset_index(drop=True)
|
24 |
+
|
25 |
+
ocr_df
|
26 |
+
|
27 |
+
ocr_df['X1']=ocr_df['left']
|
28 |
+
|
29 |
+
ocr_df['Y1']=ocr_df['top']
|
30 |
+
|
31 |
+
ocr_df['X2']= ocr_df['left'] + ocr_df['width']
|
32 |
+
|
33 |
+
ocr_df['Y2']= ocr_df['top'] + ocr_df['height']
|
34 |
+
|
35 |
+
return ocr_df
|
36 |
+
|
37 |
+
|
38 |
+
def image_to_text(image):
|
39 |
+
ocr_df= ocr_df_using_pytesseract(image)
|
40 |
+
|
41 |
+
|
42 |
+
grouped_text = ocr_df.groupby(['block_num', 'line_num'])['text'].agg(' '.join).reset_index()
|
43 |
+
|
44 |
+
# sort the text by line numbers within each block
|
45 |
+
grouped_text = grouped_text.sort_values(['block_num', 'line_num'])
|
46 |
+
|
47 |
+
# join the text by blocks and add newlines
|
48 |
+
result = ''
|
49 |
+
for i, row in grouped_text.iterrows():
|
50 |
+
if i > 0 and row['block_num'] != grouped_text.loc[i-1, 'block_num']:
|
51 |
+
result += '\n\n'
|
52 |
+
result += row['text'].rstrip() + '\n'
|
53 |
+
|
54 |
+
return result
|
55 |
+
|
56 |
+
|
57 |
+
|
58 |
+
demo = gr.Interface(fn=image_to_text,
|
59 |
+
inputs= gr.Image(type="pil"),
|
60 |
+
outputs=["text"],
|
61 |
+
title="Menu image to text",
|
62 |
+
description= "Upload Menu image here")
|
63 |
+
demo.launch(share=True)
|