File size: 3,853 Bytes
a89d9fd 17e862b 5e7197d 6c21fbb a89d9fd e3c89f9 a89d9fd 17e862b a89d9fd 17e862b c883606 17e862b c883606 5e7197d a89d9fd e3c89f9 23283c0 6c21fbb e3c89f9 c883606 e3c89f9 c883606 6c21fbb c883606 e3c89f9 6c21fbb c883606 6c21fbb e3c89f9 c883606 |
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 |
import os, io
from paddleocr import PaddleOCR, draw_ocr,PPStructure
from ppocr.utils.visual import draw_ser_results
from PIL import Image
import gradio as gr
example_dir = './example_images'
example_images = [file for file in os.listdir(example_dir) if file.endswith(('.png', '.jpg', '.jpeg'))]
def load_image(filename):
# Construct full file path
file_path = os.path.join(example_dir, filename)
# Load and return the image
return Image.open(file_path)
def inference__ppocr(img_path):
ocr = PaddleOCR(
rec_char_dict_path='zhtw_common_dict.txt',
use_gpu=False,
rec_image_shape="3, 48, 320"
)
result = ocr.ocr(img_path)
for idx in range(len(result)):
res = result[idx]
for line in res:
print(line)
result = result[0]
image = Image.open(img_path).convert('RGB')
boxes = [line[0] for line in result]
txts = [line[1][0] if line[1] else '' for line in result] # 確保在無文字時 txts 還是個空字串
scores = [line[1][1] for line in result]
im_show_pil = draw_ocr(image, boxes, txts, scores, font_path="./simfang.ttf")
return im_show_pil, "\n".join(txts)
def inference__ppstructure(img_path):
ppsutructure = PPStructure(
rec_char_dict_path='zhtw_common_dict.txt',
use_gpu=False,
rec_image_shape="3, 48, 320",
ser_dict_path='ppocr/utils/dict/kie/clinical_class_list.txt'
)
samples = ['病歷','身份','姓名',' Medical','No.','Name','性別','中華民國','002480','身分','Attending','M.D','ID','Medical','by','續上頁診斷書內容','出生地','列印時間','以上','年齡','特予']
result,_ = ppsutructure.__call__(img_path)
for element in result:
for sample in samples:
if sample in element['transcription']:
element['pred_id'] = 0
element['pred'] ='O'
image = draw_ser_results(img_path,result,font_path='./simfang.ttf')
result = [''.join(f"{element['pred']}:{element['transcription']}") for element in result if element['pred']!='O']
return image, "\n".join(result)
def update_image_input(filename):
# Construct full file path
file_path = os.path.join(example_dir, filename)
# Return the file path
return file_path
with gr.Blocks() as demo:
gr.Markdown("<p style='text-align: center; font-size: 50px; font-weight: bold;'>Form Understanding Project - Certificate of Diagnosis</p>")
gr.Markdown("Support languages: Traditional Chinese 🇹🇼")
gr.Markdown("version:0.1")
gr.Markdown("""
## Usage Description
This interface is designed to process and extract information from Certificates of Diagnosis.
To use this tool:
1. Upload an image of a Certificate of Diagnosis using the 'Upload Image' button.
3. Click 'Process' to extract information from the uploaded certificate.
4. The processed image and extracted text will be displayed on the right.
""")
with gr.Row():
with gr.Column():
gr.Markdown("#### Input Image")
image_input = gr.Image(type='filepath', label='Upload Image')
image_selection = gr.Radio(label="Select an Example Image", choices=example_images)
submit_btn = gr.Button("Process")
with gr.Column():
gr.Markdown("#### Processed Image")
image_output = gr.Image(type="pil", label="Processed Image")
gr.Markdown("#### Extracted Text")
text_output = gr.Textbox(label="Extracted Text")
image_selection.change(
update_image_input,
inputs=[image_selection],
outputs=[image_input]
)
submit_btn.click(
inference__ppstructure,
inputs=[image_input],
outputs=[image_output, text_output]
)
demo.launch(debug=True) |