import gradio as gr
from ultralytics import YOLO
import cv2
import numpy as np
from PIL import Image, ImageDraw, ImageFont
import base64
from io import BytesIO
import tempfile
import os
from pathlib import Path
import shutil
# Load YOLOv8 model
model = YOLO("best.pt")
# Create directories if not present
uploaded_folder = Path('Uploaded_Picture')
predicted_folder = Path('Predicted_Picture')
uploaded_folder.mkdir(parents=True, exist_ok=True)
predicted_folder.mkdir(parents=True, exist_ok=True)
# Path for HTML database file
html_db_file = Path('patient_predictions.html')
# Initialize HTML database file if not present
if not html_db_file.exists():
with open(html_db_file, 'w') as f:
f.write("Patient Prediction Database
")
def predict_image(input_image, name, age, medical_record, sex):
if input_image is None:
return None, "Please Input The Image"
# Convert Gradio input image (PIL Image) to numpy array
image_np = np.array(input_image)
# Ensure the image is in the correct format
if len(image_np.shape) == 2: # grayscale to RGB
image_np = cv2.cvtColor(image_np, cv2.COLOR_GRAY2RGB)
elif image_np.shape[2] == 4: # RGBA to RGB
image_np = cv2.cvtColor(image_np, cv2.COLOR_RGBA2RGB)
# Perform prediction
results = model(image_np)
# Draw bounding boxes on the image
image_with_boxes = image_np.copy()
raw_predictions = []
if results[0].boxes:
# Sort the results by confidence and take the highest confidence one
highest_confidence_result = max(results[0].boxes, key=lambda x: x.conf.item())
# Determine the label based on the class index
class_index = highest_confidence_result.cls.item()
if class_index == 0:
label = "Immature"
color = (0, 255, 255) # Yellow for Immature
elif class_index == 1:
label = "Mature"
color = (255, 0, 0) # Red for Mature
else:
label = "Normal"
color = (0, 255, 0) # Green for Normal
confidence = highest_confidence_result.conf.item()
xmin, ymin, xmax, ymax = map(int, highest_confidence_result.xyxy[0])
# Draw the bounding box
cv2.rectangle(image_with_boxes, (xmin, ymin), (xmax, ymax), color, 2)
# Enlarge font scale and thickness
font_scale = 1.0
thickness = 2
# Calculate label background size
(text_width, text_height), baseline = cv2.getTextSize(f'{label} {confidence:.2f}', cv2.FONT_HERSHEY_SIMPLEX, font_scale, thickness)
cv2.rectangle(image_with_boxes, (xmin, ymin - text_height - baseline), (xmin + text_width, ymin), (0, 0, 0), cv2.FILLED)
# Put the label text with black background
cv2.putText(image_with_boxes, f'{label} {confidence:.2f}', (xmin, ymin - 10), cv2.FONT_HERSHEY_SIMPLEX, font_scale, (255, 255, 255), thickness)
raw_predictions.append(f"Label: {label}, Confidence: {confidence:.2f}, Box: [{xmin}, {ymin}, {xmax}, {ymax}]")
raw_predictions_str = "\n".join(raw_predictions)
# Convert to PIL image for further processing
pil_image_with_boxes = Image.fromarray(image_with_boxes)
# Add text and watermark
pil_image_with_boxes = add_text_and_watermark(pil_image_with_boxes, name, age, medical_record, sex, label)
# Save images to directories
image_name = f"{name}-{age}-{sex}-{medical_record}.png"
input_image.save(uploaded_folder / image_name)
pil_image_with_boxes.save(predicted_folder / image_name)
# Convert the predicted image to base64 for embedding in HTML
buffered = BytesIO()
pil_image_with_boxes.save(buffered, format="PNG")
predicted_image_base64 = base64.b64encode(buffered.getvalue()).decode()
# Append the prediction to the HTML database
append_patient_info_to_html(name, age, medical_record, sex, label, predicted_image_base64)
return pil_image_with_boxes, raw_predictions_str
# Function to add watermark
def add_watermark(image):
try:
logo = Image.open('image-logo.png').convert("RGBA")
image = image.convert("RGBA")
# Resize logo
basewidth = 100
wpercent = (basewidth / float(logo.size[0]))
hsize = int((float(wpercent) * logo.size[1]))
logo = logo.resize((basewidth, hsize), Image.LANCZOS)
# Position logo
position = (image.width - logo.width - 10, image.height - logo.height - 10)
# Composite image
transparent = Image.new('RGBA', (image.width, image.height), (0, 0, 0, 0))
transparent.paste(image, (0, 0))
transparent.paste(logo, position, mask=logo)
return transparent.convert("RGB")
except Exception as e:
print(f"Error adding watermark: {e}")
return image
# Function to add text and watermark
def add_text_and_watermark(image, name, age, medical_record, sex, label):
draw = ImageDraw.Draw(image)
# Load a larger font (adjust the size as needed)
font_size = 24 # Example font size
try:
font = ImageFont.truetype("font.ttf", size=font_size)
except IOError:
font = ImageFont.load_default()
print("Error: cannot open resource, using default font.")
text = f"Name: {name}, Age: {age}, Medical Record: {medical_record}, Sex: {sex}, Result: {label}"
# Calculate text bounding box
text_bbox = draw.textbbox((0, 0), text, font=font)
text_width, text_height = text_bbox[2] - text_bbox[0], text_bbox[3] - text_bbox[1]
text_x = 20
text_y = 40
padding = 10
# Draw a filled rectangle for the background
draw.rectangle(
[text_x - padding, text_y - padding, text_x + text_width + padding, text_y + text_height + padding],
fill="black"
)
# Draw text on top of the rectangle
draw.text((text_x, text_y), text, fill=(255, 255, 255, 255), font=font)
# Add watermark to the image
image_with_watermark = add_watermark(image)
return image_with_watermark
def append_patient_info_to_html(name, age, medical_record, sex, result, predicted_image_base64):
# Check if the HTML file is empty or if the table structure is missing
if os.stat(html_db_file).st_size == 0: # Empty file, create the table structure
with open(html_db_file, 'a') as f:
f.write("""
Patient Prediction Database
Patient Prediction Database
| Name |
Age |
Medical Record |
Sex |
Result |
Predicted Image |
""")
# Check if this patient already exists to prevent duplicate entries
# This can be improved by checking unique identifiers like `medical_record`
# Assuming the uniqueness of the medical record
html_entry = f"""
| {name} |
{age} |
{medical_record} |
{sex} |
{result} |
 |
"""
with open(html_db_file, 'a') as f:
f.write(html_entry)
# Ensure we only add the closing tags once
if "
" not in open(html_db_file).read():
with open(html_db_file, 'a') as f:
f.write("""
""")
return str(html_db_file) # Return the HTML file path for download
# Function to download the folders
def download_folder(folder):
zip_path = os.path.join(tempfile.gettempdir(), f"{folder}.zip")
# Zip the folder
shutil.make_archive(zip_path.replace('.zip', ''), 'zip', folder)
return zip_path
# Gradio Interface
def interface(name, age, medical_record, sex, input_image):
if input_image is None:
return None, "Please upload an image.", None
output_image, raw_result = predict_image(input_image, name, age, medical_record, sex)
# Return the current state of the HTML file with all predictions
return output_image, raw_result, str(html_db_file)
# Download Functions
def download_predicted_folder():
return download_folder(predicted_folder)
def download_uploaded_folder():
return download_folder(uploaded_folder)
# Launch Gradio Interface
with gr.Blocks() as demo:
with gr.Column():
gr.Markdown("# Cataract Detection System")
gr.Markdown("Upload an image to detect cataract and add patient details.")
gr.Markdown("This application uses YOLOv8 with mAP=0.981")
with gr.Column():
name = gr.Textbox(label="Name")
age = gr.Number(label="Age")
medical_record = gr.Number(label="Medical Record")
sex = gr.Radio(["Male", "Female"], label="Sex")
input_image = gr.Image(type="pil", label="Upload an Image", image_mode="RGB")
with gr.Column():
submit_btn = gr.Button("Submit")
output_image = gr.Image(type="pil", label="Predicted Image")
with gr.Row():
raw_result = gr.Textbox(label="Prediction Result")
with gr.Row():
download_html_btn = gr.Button("Download Patient Information (HTML)")
download_uploaded_btn = gr.Button("Download Uploaded Images")
download_predicted_btn = gr.Button("Download Predicted Images")
# Add file download output components for the uploaded and predicted images
patient_info_file = gr.File(label="Patient Information HTML File")
uploaded_folder_file = gr.File(label="Uploaded Images Zip File")
predicted_folder_file = gr.File(label="Predicted Images Zip File")
# Connect functions with components
submit_btn.click(fn=interface, inputs=[name, age, medical_record, sex, input_image], outputs=[output_image, raw_result])
download_html_btn.click(fn=append_patient_info_to_html, inputs=[name, age, medical_record, sex, raw_result], outputs=patient_info_file)
download_uploaded_btn.click(fn=download_uploaded_folder, outputs=uploaded_folder_file)
download_predicted_btn.click(fn=download_predicted_folder, outputs=predicted_folder_file)
# Launch Gradio app
demo.launch()