Delete HandwritingOCR
Browse files- HandwritingOCR/.DS_Store +0 -0
- HandwritingOCR/ImagesProcessing.py +0 -42
- HandwritingOCR/OCRmodel.py +0 -138
- HandwritingOCR/app.py +0 -76
- HandwritingOCR/captured_images/captured_image.jpg +0 -0
- HandwritingOCR/captured_images/pasted_image.jpg +0 -0
- HandwritingOCR/main.py +0 -4
- HandwritingOCR/processed_images/processed_image.jpg +0 -3
- HandwritingOCR/static/css/style.css +0 -94
- HandwritingOCR/static/js/other.js +0 -93
- HandwritingOCR/static/js/script.js +0 -88
- HandwritingOCR/templates/home.html +0 -18
- HandwritingOCR/templates/index.html +0 -29
- HandwritingOCR/templates/other.html +0 -29
HandwritingOCR/.DS_Store
DELETED
Binary file (6.15 kB)
|
|
HandwritingOCR/ImagesProcessing.py
DELETED
@@ -1,42 +0,0 @@
|
|
1 |
-
import cv2
|
2 |
-
import matplotlib.pyplot as plt
|
3 |
-
from super_image import EdsrModel, ImageLoader
|
4 |
-
from PIL import Image
|
5 |
-
def preprocess_image(image_path):
|
6 |
-
img = cv2.imread(image_path)
|
7 |
-
img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
|
8 |
-
return img
|
9 |
-
def show_image(img):
|
10 |
-
plt.imshow(img, cmap='gray')
|
11 |
-
plt.axis('off')
|
12 |
-
plt.show()
|
13 |
-
def save_processed_image(img):
|
14 |
-
output_path = "Projects/HandwritingOCR/processed_images/processed_image.jpg"
|
15 |
-
cv2.imwrite(output_path, img)
|
16 |
-
return output_path
|
17 |
-
'''def createBoundingBox(img):
|
18 |
-
ocr_data = pytesseract.image_to_data(img, output_type=pytesseract.Output.DICT)
|
19 |
-
n_boxes = len(ocr_data['level'])
|
20 |
-
for i in range(n_boxes):
|
21 |
-
if ocr_data['level'][i] == 3:
|
22 |
-
(x, y, w, h) = (ocr_data['left'][i], ocr_data['top'][i], ocr_data['width'][i], ocr_data['height'][i])
|
23 |
-
cv2.rectangle(img, (x, y), (x + w, y + h), (0, 255, 0), 5)
|
24 |
-
plt.imshow(img, cmap='gray')
|
25 |
-
plt.axis('off')
|
26 |
-
plt.show()
|
27 |
-
'''
|
28 |
-
|
29 |
-
def super_resolution(img):
|
30 |
-
model = EdsrModel.from_pretrained('eugenesiow/edsr-base', scale=2)
|
31 |
-
pil_img = Image.fromarray(img)
|
32 |
-
inputs = ImageLoader.load_image(pil_img)
|
33 |
-
preds = model(inputs)
|
34 |
-
|
35 |
-
ImageLoader.save_image(preds, 'Projects/HandwritingOCR/processed_images/processed_image.jpg')
|
36 |
-
def process_image(image_path):
|
37 |
-
img = preprocess_image(image_path)
|
38 |
-
super_resolution(img)
|
39 |
-
|
40 |
-
if __name__ == "__main__":
|
41 |
-
image_path = "Projects/HandwritingOCR/captured_images/captured_image.jpg"
|
42 |
-
process_image(image_path)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
HandwritingOCR/OCRmodel.py
DELETED
@@ -1,138 +0,0 @@
|
|
1 |
-
import warnings
|
2 |
-
from urllib3.exceptions import NotOpenSSLWarning
|
3 |
-
|
4 |
-
warnings.filterwarnings("ignore", category=NotOpenSSLWarning)
|
5 |
-
warnings.filterwarnings("ignore", category=FutureWarning)
|
6 |
-
warnings.filterwarnings("ignore", category=UserWarning, module='torch')
|
7 |
-
warnings.filterwarnings("ignore", category=UserWarning, module='transformers')
|
8 |
-
import os
|
9 |
-
import numpy as np
|
10 |
-
import torch
|
11 |
-
import torchvision.transforms as T
|
12 |
-
from PIL import Image
|
13 |
-
from torchvision.transforms.functional import InterpolationMode
|
14 |
-
from transformers import AutoModel, AutoTokenizer
|
15 |
-
import matplotlib.pyplot as plt
|
16 |
-
|
17 |
-
IMAGENET_MEAN = (0.485, 0.456, 0.406)
|
18 |
-
IMAGENET_STD = (0.229, 0.224, 0.225)
|
19 |
-
|
20 |
-
#model_name = "5CD-AI/Vintern-1B-v2"
|
21 |
-
model_name = "5CD-AI/Vintern-1B-v3_5"
|
22 |
-
device = torch.device("mps" if torch.backends.mps.is_available() else "cpu")
|
23 |
-
|
24 |
-
def build_transform(input_size):
|
25 |
-
MEAN, STD = IMAGENET_MEAN, IMAGENET_STD
|
26 |
-
transform = T.Compose([
|
27 |
-
T.Lambda(lambda img: img.convert('RGB') if img.mode != 'RGB' else img),
|
28 |
-
T.Resize((input_size, input_size), interpolation=InterpolationMode.BICUBIC),
|
29 |
-
T.ToTensor(),
|
30 |
-
T.Normalize(mean=MEAN, std=STD)
|
31 |
-
])
|
32 |
-
return transform
|
33 |
-
|
34 |
-
def find_closest_aspect_ratio(aspect_ratio, target_ratios, width, height, image_size):
|
35 |
-
best_ratio_diff = float('inf')
|
36 |
-
best_ratio = (1, 1)
|
37 |
-
area = width * height
|
38 |
-
for ratio in target_ratios:
|
39 |
-
target_aspect_ratio = ratio[0] / ratio[1]
|
40 |
-
ratio_diff = abs(aspect_ratio - target_aspect_ratio)
|
41 |
-
if ratio_diff < best_ratio_diff:
|
42 |
-
best_ratio_diff = ratio_diff
|
43 |
-
best_ratio = ratio
|
44 |
-
elif ratio_diff == best_ratio_diff:
|
45 |
-
if area > 0.5 * image_size * image_size * ratio[0] * ratio[1]:
|
46 |
-
best_ratio = ratio
|
47 |
-
return best_ratio
|
48 |
-
|
49 |
-
def dynamic_preprocess(image, min_num=1, max_num=12, image_size=448, use_thumbnail=False):
|
50 |
-
orig_width, orig_height = image.size
|
51 |
-
aspect_ratio = orig_width / orig_height
|
52 |
-
|
53 |
-
# calculate the existing image aspect ratio
|
54 |
-
target_ratios = set(
|
55 |
-
(i, j) for n in range(min_num, max_num + 1) for i in range(1, n + 1) for j in range(1, n + 1) if
|
56 |
-
i * j <= max_num and i * j >= min_num)
|
57 |
-
target_ratios = sorted(target_ratios, key=lambda x: x[0] * x[1])
|
58 |
-
|
59 |
-
# find the closest aspect ratio to the target
|
60 |
-
target_aspect_ratio = find_closest_aspect_ratio(
|
61 |
-
aspect_ratio, target_ratios, orig_width, orig_height, image_size)
|
62 |
-
|
63 |
-
# calculate the target width and height
|
64 |
-
target_width = image_size * target_aspect_ratio[0]
|
65 |
-
target_height = image_size * target_aspect_ratio[1]
|
66 |
-
blocks = target_aspect_ratio[0] * target_aspect_ratio[1]
|
67 |
-
|
68 |
-
# resize the image
|
69 |
-
resized_img = image.resize((target_width, target_height))
|
70 |
-
processed_images = []
|
71 |
-
for i in range(blocks):
|
72 |
-
box = (
|
73 |
-
(i % (target_width // image_size)) * image_size,
|
74 |
-
(i // (target_width // image_size)) * image_size,
|
75 |
-
((i % (target_width // image_size)) + 1) * image_size,
|
76 |
-
((i // (target_width // image_size)) + 1) * image_size
|
77 |
-
)
|
78 |
-
# split the image
|
79 |
-
split_img = resized_img.crop(box)
|
80 |
-
processed_images.append(split_img)
|
81 |
-
assert len(processed_images) == blocks
|
82 |
-
if use_thumbnail and len(processed_images) != 1:
|
83 |
-
thumbnail_img = image.resize((image_size, image_size))
|
84 |
-
processed_images.append(thumbnail_img)
|
85 |
-
return processed_images
|
86 |
-
|
87 |
-
def load_image(image_file, input_size=448, max_num=12):
|
88 |
-
image = Image.open(image_file).convert('RGB')
|
89 |
-
transform = build_transform(input_size=input_size)
|
90 |
-
images = dynamic_preprocess(image, image_size=input_size, use_thumbnail=True, max_num=max_num)
|
91 |
-
pixel_values = [transform(image) for image in images]
|
92 |
-
pixel_values = torch.stack(pixel_values)
|
93 |
-
return pixel_values
|
94 |
-
|
95 |
-
def truncate_tokens(tokens, max_length):
|
96 |
-
if len(tokens) > max_length:
|
97 |
-
tokens = tokens[:max_length]
|
98 |
-
return tokens
|
99 |
-
|
100 |
-
def OCRing(image_URL):
|
101 |
-
test_image = image_URL
|
102 |
-
pixel_values = load_image(test_image, max_num=6).to(torch.bfloat16).to(device)
|
103 |
-
generation_config = dict(max_new_tokens=512, do_sample=False, num_beams=3, repetition_penalty=3.5)
|
104 |
-
|
105 |
-
question = '<image>\n Chỉ xuất ra kí tự có trong văn bản, không thêm bớt.'
|
106 |
-
|
107 |
-
response = model.chat(tokenizer, pixel_values, question, generation_config)
|
108 |
-
print(f'User: {question}\nAssistant: {response}')
|
109 |
-
return response
|
110 |
-
|
111 |
-
try:
|
112 |
-
model = AutoModel.from_pretrained(
|
113 |
-
model_name,
|
114 |
-
torch_dtype=torch.bfloat16,
|
115 |
-
low_cpu_mem_usage=True,
|
116 |
-
trust_remote_code=True,
|
117 |
-
use_flash_attn=False,
|
118 |
-
).eval().to(device)
|
119 |
-
except:
|
120 |
-
model = AutoModel.from_pretrained(
|
121 |
-
model_name,
|
122 |
-
torch_dtype=torch.bfloat16,
|
123 |
-
low_cpu_mem_usage=True,
|
124 |
-
trust_remote_code=True
|
125 |
-
).eval().to(device)
|
126 |
-
|
127 |
-
tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True, use_fast=False)
|
128 |
-
|
129 |
-
if __name__ == "__main__":
|
130 |
-
test_image = "Projects/HandwritingOCR/captured_images/captured_image.jpg"
|
131 |
-
pixel_values = load_image(test_image, max_num=6).to(torch.bfloat16).to(device)
|
132 |
-
generation_config = dict(max_new_tokens=512, do_sample=False, num_beams=3, repetition_penalty=3.5)
|
133 |
-
|
134 |
-
question = '<image>\n Input: ảnh, Output: Chỉ xuất ra những kí tự có trong ảnh, không thêm bớt.'
|
135 |
-
|
136 |
-
response = model.chat(tokenizer, pixel_values, question, generation_config)
|
137 |
-
print(f'User: {question}\nAssistant: {response}')
|
138 |
-
#dùng dòng lệnh này trong terminal: export PYTORCH_ENABLE_MPS_FALLBACK=1
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
HandwritingOCR/app.py
DELETED
@@ -1,76 +0,0 @@
|
|
1 |
-
from flask import Flask, render_template, request, jsonify
|
2 |
-
import cv2
|
3 |
-
import numpy as np
|
4 |
-
import os
|
5 |
-
import base64
|
6 |
-
import OCRmodel as ocr
|
7 |
-
import ImagesProcessing as ip
|
8 |
-
app = Flask(__name__)
|
9 |
-
|
10 |
-
# Tạo thư mục lưu ảnh nếu chưa có
|
11 |
-
save_dir = "Projects/HandwritingOCR/captured_images"
|
12 |
-
if not os.path.exists(save_dir):
|
13 |
-
os.makedirs(save_dir)
|
14 |
-
ocr_process = None
|
15 |
-
|
16 |
-
@app.route('/')
|
17 |
-
def home():
|
18 |
-
return render_template('home.html')
|
19 |
-
|
20 |
-
@app.route('/index')
|
21 |
-
def index():
|
22 |
-
return render_template('index.html')
|
23 |
-
|
24 |
-
@app.route('/other')
|
25 |
-
def other():
|
26 |
-
return render_template('other.html')
|
27 |
-
|
28 |
-
@app.route('/capture', methods=['POST'])
|
29 |
-
def capture():
|
30 |
-
data = request.json
|
31 |
-
image_data = data['image']
|
32 |
-
image_data = image_data.split(",")[1]
|
33 |
-
image_data = np.frombuffer(base64.b64decode(image_data), np.uint8)
|
34 |
-
image = cv2.imdecode(image_data, cv2.IMREAD_COLOR)
|
35 |
-
image_path = os.path.join(save_dir, "captured_image.jpg")
|
36 |
-
cv2.imwrite(image_path, image)
|
37 |
-
return jsonify({"message": "Image saved successfully!"})
|
38 |
-
|
39 |
-
@app.route('/save_pasted_image', methods=['POST'])
|
40 |
-
def save_pasted_image():
|
41 |
-
data = request.json
|
42 |
-
image_data = data['image']
|
43 |
-
image_data = image_data.split(",")[1]
|
44 |
-
image_data = np.frombuffer(base64.b64decode(image_data), np.uint8)
|
45 |
-
image = cv2.imdecode(image_data, cv2.IMREAD_COLOR)
|
46 |
-
image_path = os.path.join(save_dir, "pasted_image.jpg")
|
47 |
-
cv2.imwrite(image_path, image)
|
48 |
-
return jsonify({"message": "Pasted image saved successfully!"})
|
49 |
-
|
50 |
-
@app.route('/camocr', methods=['POST'])
|
51 |
-
def camocr():
|
52 |
-
image_path = os.path.join(save_dir, "captured_image.jpg")
|
53 |
-
result = DoOCR(image_path)
|
54 |
-
return jsonify({"result": result})
|
55 |
-
|
56 |
-
@app.route('/imgocr', methods=['POST'])
|
57 |
-
def imgocr():
|
58 |
-
image_path = os.path.join(save_dir, "pasted_image.jpg")
|
59 |
-
result = DoOCR(image_path)
|
60 |
-
return jsonify({"result": result})
|
61 |
-
|
62 |
-
|
63 |
-
def processImage(image_path):
|
64 |
-
ip.process_image(image_path)
|
65 |
-
#output_path = ip.save_processed_image(img)
|
66 |
-
#return output_path
|
67 |
-
|
68 |
-
def DoOCR(image_path):
|
69 |
-
processImage(image_path)
|
70 |
-
output_path = "Projects/HandwritingOCR/processed_images/processed_image.jpg"
|
71 |
-
return ocr.OCRing(output_path)
|
72 |
-
|
73 |
-
|
74 |
-
if __name__ == '__main__':
|
75 |
-
app.run(debug=True)
|
76 |
-
#dùng: export PYTORCH_ENABLE_MPS_FALLBACK=1 trong terminal
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
HandwritingOCR/captured_images/captured_image.jpg
DELETED
Binary file (58.4 kB)
|
|
HandwritingOCR/captured_images/pasted_image.jpg
DELETED
Binary file (52.3 kB)
|
|
HandwritingOCR/main.py
DELETED
@@ -1,4 +0,0 @@
|
|
1 |
-
#dùng: export PYTORCH_ENABLE_MPS_FALLBACK=1 trong terminal
|
2 |
-
import OCRmodel as ocr
|
3 |
-
image_path = '/Users/lequanhuy/Documents/Code/Visual Code/Projects/HandwritingOCR/captured_images/captured_image.jpg'
|
4 |
-
print(ocr.OCRing(image_path))
|
|
|
|
|
|
|
|
|
|
HandwritingOCR/processed_images/processed_image.jpg
DELETED
Git LFS Details
|
HandwritingOCR/static/css/style.css
DELETED
@@ -1,94 +0,0 @@
|
|
1 |
-
body {
|
2 |
-
font-family: 'Roboto', sans-serif;
|
3 |
-
background-color: #f0f0f0;
|
4 |
-
margin: 0;
|
5 |
-
padding: 0;
|
6 |
-
display: flex;
|
7 |
-
justify-content: center;
|
8 |
-
align-items: center;
|
9 |
-
height: 100%;
|
10 |
-
}
|
11 |
-
|
12 |
-
.container {
|
13 |
-
background-color: #fff;
|
14 |
-
padding: 30px;
|
15 |
-
border-radius: 10px;
|
16 |
-
box-shadow: 0 0 20px rgba(0, 0, 0, 0.1);
|
17 |
-
text-align: center;
|
18 |
-
width: 90%;
|
19 |
-
max-width: 900px;
|
20 |
-
}
|
21 |
-
|
22 |
-
h1 {
|
23 |
-
color: #333;
|
24 |
-
margin-bottom: 20px;
|
25 |
-
font-size: 2.5em;
|
26 |
-
}
|
27 |
-
|
28 |
-
p {
|
29 |
-
color: #666;
|
30 |
-
font-size: 1.2em;
|
31 |
-
}
|
32 |
-
|
33 |
-
.button-container {
|
34 |
-
margin-bottom: 20px;
|
35 |
-
}
|
36 |
-
|
37 |
-
button {
|
38 |
-
background-color: #007bff;
|
39 |
-
color: #fff;
|
40 |
-
border: none;
|
41 |
-
padding: 15px 30px;
|
42 |
-
margin: 10px;
|
43 |
-
border-radius: 5px;
|
44 |
-
cursor: pointer;
|
45 |
-
font-size: 1em;
|
46 |
-
transition: background-color 0.3s ease;
|
47 |
-
}
|
48 |
-
|
49 |
-
button:hover {
|
50 |
-
background-color: #0056b3;
|
51 |
-
}
|
52 |
-
|
53 |
-
.paste-container {
|
54 |
-
margin-top: 20px;
|
55 |
-
}
|
56 |
-
|
57 |
-
#paste-box {
|
58 |
-
border: 2px dashed #007bff;
|
59 |
-
border-radius: 5px;
|
60 |
-
padding: 20px;
|
61 |
-
min-height: 150px;
|
62 |
-
cursor: text;
|
63 |
-
background-color: #f9f9f9;
|
64 |
-
transition: background-color 0.3s ease;
|
65 |
-
}
|
66 |
-
|
67 |
-
#paste-box:focus {
|
68 |
-
background-color: #e9f7ff;
|
69 |
-
}
|
70 |
-
|
71 |
-
#pasted-image {
|
72 |
-
border: 2px solid #007bff;
|
73 |
-
border-radius: 5px;
|
74 |
-
max-width: 100%;
|
75 |
-
margin-top: 20px;
|
76 |
-
}
|
77 |
-
|
78 |
-
.ocr-result {
|
79 |
-
background-color: #e9ecef;
|
80 |
-
padding: 20px;
|
81 |
-
border-radius: 5px;
|
82 |
-
border: 1px solid #ced4da;
|
83 |
-
color: #495057;
|
84 |
-
font-size: 1em;
|
85 |
-
text-align: left;
|
86 |
-
white-space: pre-wrap;
|
87 |
-
margin-top: 20px;
|
88 |
-
}
|
89 |
-
|
90 |
-
.loading {
|
91 |
-
font-size: 1.5em;
|
92 |
-
color: #007bff;
|
93 |
-
margin-top: 20px;
|
94 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
HandwritingOCR/static/js/other.js
DELETED
@@ -1,93 +0,0 @@
|
|
1 |
-
let loadingInterval;
|
2 |
-
|
3 |
-
function handlePaste(event) {
|
4 |
-
const items = (event.clipboardData || event.originalEvent.clipboardData).items;
|
5 |
-
for (const item of items) {
|
6 |
-
if (item.type.indexOf("image") === 0) {
|
7 |
-
const blob = item.getAsFile();
|
8 |
-
const reader = new FileReader();
|
9 |
-
reader.onload = function(event) {
|
10 |
-
const img = document.getElementById("pasted-image");
|
11 |
-
img.src = event.target.result;
|
12 |
-
img.style.display = "block";
|
13 |
-
};
|
14 |
-
reader.readAsDataURL(blob);
|
15 |
-
}
|
16 |
-
}
|
17 |
-
}
|
18 |
-
|
19 |
-
function saveImage() {
|
20 |
-
const img = document.getElementById("pasted-image");
|
21 |
-
if (img.src) {
|
22 |
-
fetch('/save_pasted_image', {
|
23 |
-
method: 'POST',
|
24 |
-
headers: {
|
25 |
-
'Content-Type': 'application/json'
|
26 |
-
},
|
27 |
-
body: JSON.stringify({ image: img.src })
|
28 |
-
})
|
29 |
-
.then(response => response.json())
|
30 |
-
.then(data => {
|
31 |
-
console.log(data.message);
|
32 |
-
alert("Image saved successfully!");
|
33 |
-
})
|
34 |
-
.catch(console.error);
|
35 |
-
} else {
|
36 |
-
alert("No image to save!");
|
37 |
-
}
|
38 |
-
}
|
39 |
-
|
40 |
-
function performOCR() {
|
41 |
-
clearOCRResult();
|
42 |
-
showLoading();
|
43 |
-
disableButton();
|
44 |
-
fetch('/imgocr', {
|
45 |
-
method: 'POST',
|
46 |
-
headers: {
|
47 |
-
'Content-Type': 'application/json'
|
48 |
-
}
|
49 |
-
})
|
50 |
-
.then(response => response.json())
|
51 |
-
.then(data => {
|
52 |
-
document.getElementById('ocr-result').innerText = data.result;
|
53 |
-
hideLoading();
|
54 |
-
enableButton();
|
55 |
-
})
|
56 |
-
.catch(error => {
|
57 |
-
console.error(error);
|
58 |
-
hideLoading();
|
59 |
-
enableButton();
|
60 |
-
});
|
61 |
-
}
|
62 |
-
|
63 |
-
function showLoading() {
|
64 |
-
const loadingElement = document.getElementById('loading');
|
65 |
-
loadingElement.style.display = 'block';
|
66 |
-
let dots = 0;
|
67 |
-
loadingInterval = setInterval(() => {
|
68 |
-
dots = (dots + 1) % 4;
|
69 |
-
loadingElement.innerText = 'Loading' + '.'.repeat(dots);
|
70 |
-
}, 500);
|
71 |
-
}
|
72 |
-
|
73 |
-
function hideLoading() {
|
74 |
-
clearInterval(loadingInterval);
|
75 |
-
const loadingElement = document.getElementById('loading');
|
76 |
-
loadingElement.style.display = 'none';
|
77 |
-
}
|
78 |
-
|
79 |
-
function clearOCRResult() {
|
80 |
-
document.getElementById('ocr-result').innerText = '';
|
81 |
-
}
|
82 |
-
|
83 |
-
function disableButton() {
|
84 |
-
const button = document.querySelector('button[onclick="performOCR()"]');
|
85 |
-
button.disabled = true;
|
86 |
-
button.style.backgroundColor = '#cccccc';
|
87 |
-
}
|
88 |
-
|
89 |
-
function enableButton() {
|
90 |
-
const button = document.querySelector('button[onclick="performOCR()"]');
|
91 |
-
button.disabled = false;
|
92 |
-
button.style.backgroundColor = '#007bff';
|
93 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
HandwritingOCR/static/js/script.js
DELETED
@@ -1,88 +0,0 @@
|
|
1 |
-
let video;
|
2 |
-
|
3 |
-
function startWebcam() {
|
4 |
-
video = document.getElementById('webcam');
|
5 |
-
navigator.mediaDevices.getUserMedia({ video: true })
|
6 |
-
.then(stream => {
|
7 |
-
video.srcObject = stream;
|
8 |
-
})
|
9 |
-
.catch(console.error);
|
10 |
-
}
|
11 |
-
|
12 |
-
function captureImage() {
|
13 |
-
const canvas = document.createElement('canvas');
|
14 |
-
canvas.width = video.videoWidth;
|
15 |
-
canvas.height = video.videoHeight;
|
16 |
-
const context = canvas.getContext('2d');
|
17 |
-
context.drawImage(video, 0, 0, canvas.width, canvas.height);
|
18 |
-
const imageData = canvas.toDataURL('image/jpeg');
|
19 |
-
fetch('/capture', {
|
20 |
-
method: 'POST',
|
21 |
-
headers: {
|
22 |
-
'Content-Type': 'application/json'
|
23 |
-
},
|
24 |
-
body: JSON.stringify({ image: imageData })
|
25 |
-
})
|
26 |
-
.then(response => response.json())
|
27 |
-
.then(data => {
|
28 |
-
console.log(data.message);
|
29 |
-
document.getElementById('captured-image').src = imageData;
|
30 |
-
document.getElementById('captured-image').style.display = 'block';
|
31 |
-
})
|
32 |
-
.catch(console.error);
|
33 |
-
}
|
34 |
-
|
35 |
-
function performOCR() {
|
36 |
-
clearOCRResult();
|
37 |
-
showLoading();
|
38 |
-
disableButton();
|
39 |
-
fetch('/camocr', {
|
40 |
-
method: 'POST',
|
41 |
-
headers: {
|
42 |
-
'Content-Type': 'application/json'
|
43 |
-
}
|
44 |
-
})
|
45 |
-
.then(response => response.json())
|
46 |
-
.then(data => {
|
47 |
-
document.getElementById('ocr-result').innerText = data.result;
|
48 |
-
hideLoading();
|
49 |
-
enableButton();
|
50 |
-
})
|
51 |
-
.catch(error => {
|
52 |
-
console.error(error);
|
53 |
-
hideLoading();
|
54 |
-
enableButton();
|
55 |
-
});
|
56 |
-
}
|
57 |
-
|
58 |
-
function showLoading() {
|
59 |
-
const loadingElement = document.getElementById('loading');
|
60 |
-
loadingElement.style.display = 'block';
|
61 |
-
let dots = 0;
|
62 |
-
loadingInterval = setInterval(() => {
|
63 |
-
dots = (dots + 1) % 4;
|
64 |
-
loadingElement.innerText = 'Loading' + '.'.repeat(dots);
|
65 |
-
}, 500);
|
66 |
-
}
|
67 |
-
|
68 |
-
function hideLoading() {
|
69 |
-
clearInterval(loadingInterval);
|
70 |
-
const loadingElement = document.getElementById('loading');
|
71 |
-
loadingElement.style.display = 'none';
|
72 |
-
}
|
73 |
-
|
74 |
-
function clearOCRResult() {
|
75 |
-
document.getElementById('ocr-result').innerText = '';
|
76 |
-
}
|
77 |
-
|
78 |
-
function disableButton() {
|
79 |
-
const button = document.querySelector('button[onclick="performOCR()"]');
|
80 |
-
button.disabled = true;
|
81 |
-
button.style.backgroundColor = '#cccccc';
|
82 |
-
}
|
83 |
-
|
84 |
-
function enableButton() {
|
85 |
-
const button = document.querySelector('button[onclick="performOCR()"]');
|
86 |
-
button.disabled = false;
|
87 |
-
button.style.backgroundColor = '#007bff';
|
88 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
HandwritingOCR/templates/home.html
DELETED
@@ -1,18 +0,0 @@
|
|
1 |
-
<!DOCTYPE html>
|
2 |
-
<html lang="en">
|
3 |
-
<head>
|
4 |
-
<meta charset="UTF-8">
|
5 |
-
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
6 |
-
<title>Home</title>
|
7 |
-
<link rel="stylesheet" href="/static/css/style.css">
|
8 |
-
</head>
|
9 |
-
<body>
|
10 |
-
<div class="container">
|
11 |
-
<h1>Welcome to VN OCR</h1>
|
12 |
-
<div class="button-container">
|
13 |
-
<button type="button" onclick="window.location.href='/index'">Go to Webcam Capture</button>
|
14 |
-
<button type="button" onclick="window.location.href='/other'">Go to Image OCR</button>
|
15 |
-
</div>
|
16 |
-
</div>
|
17 |
-
</body>
|
18 |
-
</html>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
HandwritingOCR/templates/index.html
DELETED
@@ -1,29 +0,0 @@
|
|
1 |
-
<!DOCTYPE html>
|
2 |
-
<html lang="en">
|
3 |
-
<head>
|
4 |
-
<meta charset="UTF-8">
|
5 |
-
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
6 |
-
<title>Webcam Capture</title>
|
7 |
-
<link rel="stylesheet" href="/static/css/style.css">
|
8 |
-
</head>
|
9 |
-
<body>
|
10 |
-
<div class="container">
|
11 |
-
<h1>VN OCR</h1>
|
12 |
-
<div class="button-container">
|
13 |
-
<button type="button" onclick="startWebcam()">Start Webcam</button>
|
14 |
-
<button type="button" onclick="captureImage()">Capture Image</button>
|
15 |
-
<button type="button" onclick="performOCR()">Perform OCR</button>
|
16 |
-
<button type="button" onclick="window.location.href='/'">Go to Home</button>
|
17 |
-
</div>
|
18 |
-
<div id="webcam-container">
|
19 |
-
<video id="webcam" autoplay playsinline width="640" height="480"></video>
|
20 |
-
</div>
|
21 |
-
<div id="image-container">
|
22 |
-
<img id="captured-image" src="" alt="Captured Image" style="display: none;">
|
23 |
-
</div>
|
24 |
-
<div id="ocr-result" class="ocr-result"></div>
|
25 |
-
<div id="loading" class="loading" style="display: none;">Loading</div>
|
26 |
-
</div>
|
27 |
-
<script src="/static/js/script.js"></script>
|
28 |
-
</body>
|
29 |
-
</html>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
HandwritingOCR/templates/other.html
DELETED
@@ -1,29 +0,0 @@
|
|
1 |
-
<!DOCTYPE html>
|
2 |
-
<html lang="en">
|
3 |
-
<head>
|
4 |
-
<meta charset="UTF-8">
|
5 |
-
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
6 |
-
<title>Image OCR</title>
|
7 |
-
<link rel="stylesheet" href="/static/css/style.css">
|
8 |
-
</head>
|
9 |
-
<body>
|
10 |
-
<div class="container">
|
11 |
-
<h1>Image OCR</h1>
|
12 |
-
<div class="button-container">
|
13 |
-
<button type="button" onclick="window.location.href='/'">Go to Home</button>
|
14 |
-
<button type="button" onclick="performOCR()">Perform OCR</button>
|
15 |
-
</div>
|
16 |
-
<div class="paste-container">
|
17 |
-
<h2>Paste your image here</h2>
|
18 |
-
<div id="paste-box" contenteditable="true" onpaste="handlePaste(event)">
|
19 |
-
<p></p>
|
20 |
-
</div>
|
21 |
-
<img id="pasted-image" src="" alt="Pasted Image" style="display: none;">
|
22 |
-
<button type="button" onclick="saveImage()">Save Image</button>
|
23 |
-
</div>
|
24 |
-
<div id="ocr-result" class="ocr-result"></div>
|
25 |
-
<div id="loading" class="loading" style="display: none;">Loading</div>
|
26 |
-
</div>
|
27 |
-
<script src="/static/js/other.js"></script>
|
28 |
-
</body>
|
29 |
-
</html>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|