Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
@@ -1,144 +1,120 @@
|
|
|
|
|
|
|
|
|
|
|
|
1 |
import os
|
|
|
2 |
import numpy as np
|
3 |
-
import
|
4 |
-
import
|
5 |
-
import
|
6 |
-
import re as r
|
7 |
-
from urllib.request import urlopen
|
8 |
from datetime import datetime
|
9 |
-
|
|
|
|
|
|
|
10 |
from PIL import Image
|
11 |
-
from
|
12 |
-
|
13 |
-
|
14 |
-
|
15 |
-
|
16 |
-
|
17 |
-
|
18 |
-
|
19 |
-
|
20 |
-
|
21 |
-
|
22 |
-
os.makedirs(LOCAL_DIR,exist_ok=True)
|
23 |
-
|
24 |
-
try:
|
25 |
-
hf_hub_download(
|
26 |
-
repo_id=DATASET_REPO_ID,
|
27 |
-
filename=DATA_FILENAME,
|
28 |
-
cache_dir=DATA_DIRNAME,
|
29 |
-
force_filename=DATA_FILENAME
|
30 |
-
)
|
31 |
-
|
32 |
-
except:
|
33 |
-
print("file not found")
|
34 |
|
35 |
-
|
36 |
-
|
37 |
-
|
38 |
-
|
39 |
-
print("Error occurred during git pull:", e)
|
40 |
|
41 |
-
|
42 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
43 |
|
44 |
-
|
45 |
-
|
46 |
-
|
47 |
-
|
48 |
-
|
49 |
-
|
50 |
-
return r.compile(r'Address: (\d+\.\d+\.\d+\.\d+)').search(d).group(1)
|
51 |
-
except Exception as e:
|
52 |
-
print("Error while getting IP address -->",e)
|
53 |
-
return ip_address
|
54 |
|
55 |
-
|
56 |
-
|
57 |
-
|
58 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
59 |
|
60 |
-
|
61 |
-
|
62 |
-
|
63 |
-
|
64 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
65 |
|
66 |
-
|
67 |
-
|
68 |
-
|
|
|
69 |
|
70 |
-
response = requests.request("POST", url, headers=headers, data=json.dumps(req_data))
|
71 |
-
response = response.json()
|
72 |
-
print("response======>>",response)
|
73 |
-
return response
|
74 |
-
except Exception as e:
|
75 |
-
print("Error while getting location -->",e)
|
76 |
-
return location
|
77 |
|
78 |
"""
|
79 |
-
|
80 |
"""
|
81 |
-
def dump_json(thing,file):
|
82 |
-
with open(file,'w+',encoding="utf8") as f:
|
83 |
-
json.dump(thing,f)
|
84 |
|
85 |
-
|
86 |
-
|
87 |
-
|
88 |
-
|
89 |
-
adversarial_number = 0
|
90 |
-
adversarial_number = 0 if None else adversarial_number
|
91 |
|
92 |
-
|
93 |
-
|
94 |
-
|
95 |
-
|
|
|
|
|
|
|
|
|
|
|
96 |
|
97 |
-
metadata_name = datetime.now().strftime('%Y-%m-%d %H-%M-%S')
|
98 |
-
SAVE_FILE_DIR = os.path.join(LOCAL_DIR,metadata_name)
|
99 |
-
os.makedirs(SAVE_FILE_DIR,exist_ok=True)
|
100 |
-
image_output_filename = os.path.join(SAVE_FILE_DIR,'image.png')
|
101 |
-
print("image_output_filename :",image_output_filename)
|
102 |
-
print(input_image)
|
103 |
-
try:
|
104 |
-
Image.fromarray(input_image).save(image_output_filename)
|
105 |
-
# input_image.save(image_output_filename)
|
106 |
-
except Exception:
|
107 |
-
raise Exception(f"Had issues saving np array image to file")
|
108 |
|
109 |
-
|
110 |
-
|
111 |
-
|
112 |
-
'generated_text':text_output,'ip':ip_address, 'location':location
|
113 |
-
}
|
114 |
-
|
115 |
-
dump_json(metadata,json_file_path)
|
116 |
-
|
117 |
-
# Simply upload the image file and metadata using the hub's upload_file
|
118 |
-
# Upload the image
|
119 |
-
repo_image_path = os.path.join(REPOSITORY_DIR,os.path.join(metadata_name,'image.png'))
|
120 |
-
|
121 |
-
_ = upload_file(path_or_fileobj = image_output_filename,
|
122 |
-
path_in_repo =repo_image_path,
|
123 |
-
repo_id=DATASET_REPO_ID,
|
124 |
-
repo_type='dataset',
|
125 |
-
token=HF_TOKEN
|
126 |
-
)
|
127 |
-
|
128 |
-
# Upload the metadata
|
129 |
-
repo_json_path = os.path.join(REPOSITORY_DIR,os.path.join(metadata_name,'metadata.jsonl'))
|
130 |
-
_ = upload_file(path_or_fileobj = json_file_path,
|
131 |
-
path_in_repo =repo_json_path,
|
132 |
-
repo_id= DATASET_REPO_ID,
|
133 |
-
repo_type='dataset',
|
134 |
-
token=HF_TOKEN
|
135 |
-
)
|
136 |
-
adversarial_number+=1
|
137 |
-
repo.git_pull()
|
138 |
-
|
139 |
-
url = 'http://pragnakalpdev35.pythonanywhere.com/HF_space_image_to_text'
|
140 |
-
myobj = {'Method': Method,'text_output':text_output,'img':input_image.tolist(),'ip_address':ip_address, 'loc':location}
|
141 |
-
x = requests.post(url, json = myobj)
|
142 |
-
print("mail status code",x.status_code)
|
143 |
-
|
144 |
-
return "*****Logs save successfully!!!!"
|
|
|
1 |
+
import gradio as gr
|
2 |
+
import tensorflow as tf
|
3 |
+
import keras_ocr
|
4 |
+
import requests
|
5 |
+
import cv2
|
6 |
import os
|
7 |
+
import csv
|
8 |
import numpy as np
|
9 |
+
import pandas as pd
|
10 |
+
import huggingface_hub
|
11 |
+
from huggingface_hub import Repository
|
|
|
|
|
12 |
from datetime import datetime
|
13 |
+
import scipy.ndimage.interpolation as inter
|
14 |
+
import easyocr
|
15 |
+
import datasets
|
16 |
+
from datasets import load_dataset, Image
|
17 |
from PIL import Image
|
18 |
+
from paddleocr import PaddleOCR
|
19 |
+
from save_data import flag
|
20 |
+
|
21 |
+
"""
|
22 |
+
Paddle OCR
|
23 |
+
"""
|
24 |
+
def ocr_with_paddle(img):
|
25 |
+
finaltext = ''
|
26 |
+
ocr = PaddleOCR(lang='en', use_angle_cls=True)
|
27 |
+
# img_path = 'exp.jpeg'
|
28 |
+
result = ocr.ocr(img)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
29 |
|
30 |
+
for i in range(len(result[0])):
|
31 |
+
text = result[0][i][1][0]
|
32 |
+
finaltext += ' '+ text
|
33 |
+
return finaltext
|
|
|
34 |
|
35 |
+
"""
|
36 |
+
Keras OCR
|
37 |
+
"""
|
38 |
+
def ocr_with_keras(img):
|
39 |
+
output_text = ''
|
40 |
+
pipeline=keras_ocr.pipeline.Pipeline()
|
41 |
+
images=[keras_ocr.tools.read(img)]
|
42 |
+
predictions=pipeline.recognize(images)
|
43 |
+
first=predictions[0]
|
44 |
+
for text,box in first:
|
45 |
+
output_text += ' '+ text
|
46 |
+
return output_text
|
47 |
|
48 |
+
"""
|
49 |
+
easy OCR
|
50 |
+
"""
|
51 |
+
# gray scale image
|
52 |
+
def get_grayscale(image):
|
53 |
+
return cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
|
|
|
|
|
|
|
|
|
54 |
|
55 |
+
# Thresholding or Binarization
|
56 |
+
def thresholding(src):
|
57 |
+
return cv2.threshold(src,127,255, cv2.THRESH_TOZERO)[1]
|
58 |
+
def ocr_with_easy(img):
|
59 |
+
gray_scale_image=get_grayscale(img)
|
60 |
+
thresholding(gray_scale_image)
|
61 |
+
cv2.imwrite('image.png',gray_scale_image)
|
62 |
+
reader = easyocr.Reader(['th','en'])
|
63 |
+
bounds = reader.readtext('image.png',paragraph="False",detail = 0)
|
64 |
+
bounds = ''.join(bounds)
|
65 |
+
return bounds
|
66 |
+
|
67 |
+
"""
|
68 |
+
Generate OCR
|
69 |
+
"""
|
70 |
+
def generate_ocr(Method,img):
|
71 |
|
72 |
+
text_output = ''
|
73 |
+
if (img).any():
|
74 |
+
add_csv = []
|
75 |
+
image_id = 1
|
76 |
+
print("Method___________________",Method)
|
77 |
+
if Method == 'EasyOCR':
|
78 |
+
text_output = ocr_with_easy(img)
|
79 |
+
if Method == 'KerasOCR':
|
80 |
+
text_output = ocr_with_keras(img)
|
81 |
+
if Method == 'PaddleOCR':
|
82 |
+
text_output = ocr_with_paddle(img)
|
83 |
+
|
84 |
+
try:
|
85 |
+
flag(Method,text_output,img)
|
86 |
+
except Exception as e:
|
87 |
+
print(e)
|
88 |
+
return text_output
|
89 |
+
else:
|
90 |
+
raise gr.Error("Please upload an image!!!!")
|
91 |
|
92 |
+
# except Exception as e:
|
93 |
+
# print("Error in ocr generation ==>",e)
|
94 |
+
# text_output = "Something went wrong"
|
95 |
+
# return text_output
|
96 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
97 |
|
98 |
"""
|
99 |
+
Create user interface for OCR demo
|
100 |
"""
|
|
|
|
|
|
|
101 |
|
102 |
+
# image = gr.Image(shape=(300, 300))
|
103 |
+
image = gr.Image()
|
104 |
+
method = gr.Radio(["PaddleOCR","EasyOCR", "KerasOCR"],value="PaddleOCR")
|
105 |
+
output = gr.Textbox(label="Output")
|
|
|
|
|
106 |
|
107 |
+
demo = gr.Interface(
|
108 |
+
generate_ocr,
|
109 |
+
[method,image],
|
110 |
+
output,
|
111 |
+
title="Optical Character Recognition",
|
112 |
+
css=".gradio-container {background-color: lightgray} #radio_div {background-color: #FFD8B4; font-size: 40px;}",
|
113 |
+
article = """<p style='text-align: center;'>Feel free to give us your thoughts on this demo and please contact us at
|
114 |
+
<a href="mailto:[email protected]" target="_blank">[email protected]</a>
|
115 |
+
<p style='text-align: center;'>Developed by: <a href="https://www.pragnakalp.com" target="_blank">Pragnakalp Techlabs</a></p>"""
|
116 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
117 |
|
118 |
+
)
|
119 |
+
# demo.launch(enable_queue = False)
|
120 |
+
demo.launch()
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|