-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathmain.py
273 lines (231 loc) · 12.9 KB
/
main.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
import numpy as np
import streamlit as st
import easyocr
import cv2
import re
from PIL import Image
import gspread
from oauth2client.service_account import ServiceAccountCredentials
from time import ctime
# initalize the reader
reader = easyocr.Reader(['en'])
#https://colab.research.google.com/drive/1-MKuJNyAiUpeVZtf98ApLSzp6KTFGz-_#scrollTo=Je6ka6rrx8Wx&uniqifier=1
# Function to append a list as a row to Google Sheet
def append_to_google_sheet(data):
# Load Google Sheets API credentials from the JSON file
scope = ['https://spreadsheets.google.com/feeds', 'https://www.googleapis.com/auth/drive']
creds = ServiceAccountCredentials.from_json_keyfile_name('./idcard-405217-79628c52f4d1.json', scope)
client = gspread.authorize(creds)
# Open the Google Sheet by title
sheet = client.open('Id card ocr datas').sheet1
# Append the lsit as a row to the sheet
sheet.append_row(data)
def easyocr_predicted(img):
image_text = reader.readtext(img) # it takes 1 minutes 35 seconds based on the computation power
# print(image_text)
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
font = cv2.FONT_HERSHEY_SIMPLEX
for detection in image_text:
top_left = tuple([int(x) for x in detection[0][
0]]) # ([[682).1135155856441, 839.0994685768342], [768.8559704728243, 851.9362840074967], [761.8864844143559, 890.9005314231658], [674.1440295271757, 878.0637159925033]], 'Sat', 0.007399932870564301)]
bottom_right = tuple([int(x) for x in detection[0][2]])
text = detection[1]
# print(top_left,bottom_right)
img = cv2.rectangle(img, top_left, bottom_right, (0, 255, 0), 3)
img = cv2.putText(img, text, top_left, font, 5, (0, 0, 255), 5, cv2.LINE_AA)
# cv2_imshow(img)
return image_text, img
def main_ocr(img):
output_text, predicted_img = easyocr_predicted(img)
result = "" # an empty string
for single_output in output_text:
# structure of single_output -> ([[681, 314], [1603, 314], [1603, 527], [681, 527]], 'BANNARI', 0.9994982510354642)
# a tuple -> 0th index-coordinates(top_right,top_left,bottom_right,bottom_left)
# 1st index - predicted text
# 2nd or last index -> confidence score
top_left = single_output[0][0] # [681, 314]
bottom_left = single_output[0][-1]
top_right = single_output[0][1]
height = bottom_left[1] - top_left[1] # to find the height of the predicted segment
width = top_right[0] - top_left[0] # to find the height of the predicted segment
if height > 1000 or height > width * 2 or height > width * 1.8: # ([[278, 1059], [547, 1059], [547, 2195], [278, 2195]], '0', 0.36467499949708326):
# print(height,width)
cropped_result = post_processing(single_output, img,
rotate_option="clock") # [([[9, 0], [1133, 0], [1133, 247], [9, 247]], '2021-2025', 0.9998907497931622)]
print(cropped_result, "cc")
for i in cropped_result: # i -> ([[9, 0], [1133, 0], [1133, 247], [9, 247]], '2021-2025', 0.9998907497931622)
if i[-1] > 0.5:
result += i[1] + '\n' # '2021'
else:
cropped_result2 = post_processing(single_output, img, rotate_option="counter_clock")
print(cropped_result2, "cc")
for j in cropped_result2: # i -> ([[9, 0], [1133, 0], [1133, 247], [9, 247]], '2021-2025', 0.9998907497931622)
if j[-1] > 0.5:
result += j[1] + '\n' # '2021-2025'
elif single_output[-1] > 0.5:
print("perfect : ", single_output)
result += single_output[1] + '\n'
print(result)
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
return result, predicted_img
def post_processing(single_output, img, rotate_option):
row_start = single_output[0][0][1]
row_end = single_output[0][-1][1]
col_start = single_output[0][0][0]
col_end = single_output[0][1][0]
cropped = img[row_start:row_end, col_start:col_end] # cropping the image # [rows->y axis, columns -> x axis]
if rotate_option == 'clock':
cropped = cv2.rotate(cropped, cv2.ROTATE_90_CLOCKWISE)
elif rotate_option == 'counter_clock':
cropped = cv2.rotate(cropped, cv2.ROTATE_90_COUNTERCLOCKWISE)
# cv2_imshow(cropped)
cropped_result, pred_img = easyocr_predicted(cropped)
# cv2_imshow(pred_img)
return cropped_result
def check_empty_return_value(arr):
if arr == []:
return ""
if type(arr[0]) == str: # ['7376212al114'] -> '7' == str
return arr[0]
arr = arr[0] # due to the first index is the matched index in the group of regex
if arr ==[]:
return ""
# for i in name:
# if i!="":
# name=i
# break
return [name for name in arr if name != ""][0] # shorthand for loop that stores result in list then print the 0th index
def front_text_classification(image_front_text):
front_details_result = {'Name': "", 'Roll NO': "", "Degree": "", "Department": "", "Batch": '',
'Accomodation Type': ''}
# ------------------------------ BATCH ---------------------------------
temp_batch = re.findall(r'\d{4}-\d{4}|\d{4}\n\d{4}', image_front_text)
temp_batch = [batch for batch in temp_batch if batch != ""] # ['2027\n2023']
# print("temp_batch",temp_batch)
if temp_batch != []:
if '-' in temp_batch[0]:
front_details_result["Batch"] = temp_batch[0] # ['2021-2025'] -> 2021-2025
else:
temp_batch = temp_batch[0] # ['2025\n2021'] -> 2025\n2021
temp_batch = temp_batch.split() # ['2025','2021']
front_details_result["Batch"] = temp_batch[1] + "-" + temp_batch[0] # 2021-2025
else:
front_details_result["Batch"] = ""
# ----------------------------- NAME ----------------------------------
temp_name = re.findall(r'\d{4}-\d{4}\n(.*)|(.*)\n\d{7}\D{2}\d{3}',
image_front_text) # based on the roll no or batch
front_details_result["Name"] = check_empty_return_value(temp_name)
# ---------------------------- ROLL NO ----------------------------------
temp_roll_no = re.findall(r'\d{7}\D{2}\d{3}', image_front_text) # ['7376212al114']
front_details_result["Roll NO"] = check_empty_return_value(temp_roll_no) # 7376212al114 or ""
# ---------------------------- Degree ----------------------------------
temp_degree = re.findall(r'\D\..*\.|\D\..*:|\D\D:', image_front_text)
front_details_result["Degree"] = check_empty_return_value(temp_degree)[
:-1] # B.Tech. or B.E. or BE: or B.Tech: -> B.E or B.Tech
# ---------------------------- Department ----------------------------------
temp_department = re.findall(r'\D\..*\.\n(.*)|\D{2}:\n(.*)', image_front_text)
front_details_result["Department"] = check_empty_return_value(temp_department) # AIML -> depended on degree
# ---------------------------- Accomodation Type ----------------------------------
temp_accomodation = re.findall(r'\n(H)\n|\n(D)\n', image_front_text)
temp_accomodation = check_empty_return_value(temp_accomodation) # H or D
if temp_accomodation == 'H':
front_details_result["Accomodation Type"] = "Hosteller"
elif temp_accomodation == 'D':
front_details_result["Accomodation Type"] = "Day Scholar"
final = f"""
Name : {front_details_result["Name"]}
Roll NO : {front_details_result["Roll NO"]}
Degree : {front_details_result["Degree"]}
Department : {front_details_result["Department"]}
Batch : {front_details_result["Batch"]}
Accomodation Type : {front_details_result["Accomodation Type"]}
"""
return final, front_details_result
def back_text_classification(image_back_text):
back_details_result = {"Blood Group": "", "Date Of Birth": "", "Student Number": "", "Parent Number": "",
"Official Email": "", "Address": ""}
# ------------------------------- Date Of Birth ----------------------------
temp_dob = re.findall(r'D.O.B\n(.*)|DOB :\n(.*)|D.O.B :\n(.*)|D.O.B.\n:\n(.*)|D.OB.*\n(.*)|\d{2}-\d{2}-\d{4}', image_back_text)
back_details_result["Date Of Birth"] = check_empty_return_value(temp_dob)
# ------------------------------- Blood Group ------------------------------
temp_blood = re.findall(r'.+ve|.-ve', image_back_text)
back_details_result["Blood Group"] = check_empty_return_value(temp_blood)
# -------------------------------- Address ---------------------------------
temp_address = re.findall(r'ADDRESS\n((\n|.)*)STUDENT PHONE|ADDRESS :\n((\n|.)*)STUDENT PHONE',
image_back_text) # 0th index -> ('SIO, SELVARAJ M\n239,GURU TEX COMPLEXKA-\nLIYAMMAN KOVIL\nELAMPILLAI\nSANKAGARI\nSALEM\n637502\n', '\n')
back_details_result["Address"] = check_empty_return_value(temp_address)
# -------------------------------- Student Number --------------------------
temp_stu_num = re.findall(r'STUDENT PHONE\n(\d{10})|STUDENT PHONE :\n(\d{10})', image_back_text)
back_details_result["Student Number"] = check_empty_return_value(temp_stu_num)
# -------------------------------- Parent Number --------------------------
temp_par_num = re.findall(r'PARENT PHONE\n(\d{10})|PARENT PHONE :\n(\d{10})', image_back_text)
back_details_result["Parent Number"] = check_empty_return_value(temp_par_num)
# -------------------------------- Official Email --------------------------
temp_email = re.findall(r'.*@.*', image_back_text)
back_details_result["Official Email"] = check_empty_return_value(temp_email)
final = f"""
Blood Group : {back_details_result["Blood Group"]}
Date Of Birth : {back_details_result["Date Of Birth"]}
Student Number : {back_details_result["Student Number"]}
Parent Number : {back_details_result["Parent Number"]}
Official Email : {back_details_result["Official Email"]}
Address :
{back_details_result["Address"]}
"""
return final, back_details_result
# image_back_text=main_ocr(image_back)
if __name__ == "__main__":
# to change the title and icon
st.set_page_config(page_title="ExtraID", page_icon="🪪", layout="wide", initial_sidebar_state="collapsed")
st.markdown("<h1 style='text-align: center;'>Extract the data 📃 From ID card 🪪</h1>", unsafe_allow_html=True)
# st.title("Extract the data in ID card 🪪") # like h1 tag
col1, col2 = st.columns([0.5, 0.5])
with col1:
st.markdown('<h3 style="text-align: center;margin-top:2rem;">Front side</h3>', unsafe_allow_html=True)
image_front = st.file_uploader("Choose Front Side Image of ID card")
if image_front:
image_front = Image.open(image_front)
image_front = np.array(image_front)
with st.spinner('\t\t\t\t\tWorking on it 🧑💻...'):
image_front_text, img_front_pred = main_ocr(image_front)
img_front_pred = cv2.cvtColor(img_front_pred, cv2.COLOR_BGR2RGB)
st.image(img_front_pred)
st.subheader("Extracted Text of Front Side")
st.text(image_front_text)
st.subheader("Classified Text of Front Side")
front_show_text, front_details_result = front_text_classification(image_front_text)
st.code(front_show_text)
with col2:
st.markdown('<h3 style="text-align: center;margin-top:2rem;">Back side</h3>',
unsafe_allow_html=True)
image_back = st.file_uploader("Choose Back Side Image of ID card")
if image_back:
image_back = Image.open(image_back)
image_back = np.array(image_back)
with st.spinner('\t\t\t\t\tWorking on it 🧑💻...'):
image_back_text, img_back_pred = main_ocr(image_back)
img_back_pred = cv2.cvtColor(img_back_pred,cv2.COLOR_BGR2RGB)
st.image(img_back_pred)
st.subheader("Extracted Text of Back Side")
st.text(image_back_text)
st.subheader("Classified Text of Back Side")
back_show_text, back_details_result = back_text_classification(image_back_text)
st.code(back_show_text)
#sending the data to the gsheet
full_details_push_to_gsheet = [ctime()]
full_details_push_to_gsheet += list(front_details_result.values()) + list(back_details_result.values())
print(full_details_push_to_gsheet)
append_to_google_sheet(full_details_push_to_gsheet)
# center the button
col1, col2, col3, col4, col5 = st.columns(5)
with col1:
pass
with col2:
pass
with col4:
pass
with col5:
pass
with col3:
st.link_button("See all the data here 👉 ","https://docs.google.com/spreadsheets/d/183D8pChQlxFH1Km21KEke5olm-BmVVDgA0UFLxiY0j4/edit?usp=sharing")