app.py

import gradio as gr
import logging
from roboflow import Roboflow
from PIL import Image, ImageDraw, ImageFont, ImageFilter
import os

# Configure logging
logging.basicConfig(
    level=logging.DEBUG,
    format='%(asctime)s - %(levelname)s - %(message)s',
    handlers=[
        logging.FileHandler("debug.log"),
        logging.StreamHandler()
    ]
)

# Roboflow and model configuration
ROBOFLOW_API_KEY = "YOUR_ROBOFLOW_API_KEY"  # Replace with your API key
PROJECT_NAME = "model_verification_project"
VERSION_NUMBER = 1
FONT_PATH = "./STEVEHANDWRITING-REGULAR.TTF"

# Function to process image and overlay text
def process_image(image, text):
    try:
        # Initialize Roboflow
        rf = Roboflow(api_key=ROBOFLOW_API_KEY)
        project = rf.workspace().project(PROJECT_NAME)
        model = project.version(VERSION_NUMBER).model

        # Save input image temporarily
        input_image_path = "/tmp/input_image.jpg"
        image.save(input_image_path)

        # Perform inference
        logging.debug("Performing inference on the image...")
        prediction = model.predict(input_image_path, confidence=40, overlap=30).json()
        logging.debug(f"Inference result: {prediction}")

        # Open the image for processing
        pil_image = image.convert("RGBA")

        # Iterate over detected objects
        for obj in prediction['predictions']:
            x1 = int(obj['x'] - obj['width'] / 2)
            y1 = int(obj['y'] - obj['height'] / 2)
            x2 = int(obj['x'] + obj['width'] / 2)
            y2 = int(obj['y'] + obj['height'] / 2)

            # Calculate dynamic font size
            box_width = x2 - x1
            box_height = y2 - y1
            font_size = int(min(box_width // len(text), box_height // 2) * 0.8)

            # Load font
            try:
                font = ImageFont.truetype(FONT_PATH, size=font_size)
            except Exception as e:
                logging.warning(f"Error loading font. Using default. {e}")
                font = ImageFont.load_default()

            # Calculate text position
            text_layer = Image.new("RGBA", pil_image.size, (255, 255, 255, 0))
            text_draw = ImageDraw.Draw(text_layer)

            text_width, text_height = text_draw.textsize(text, font=font)
            text_x = x1 + (box_width - text_width) // 2
            text_y = y1 + (box_height - text_height) // 2

            # Draw text on a transparent layer
            text_draw.text((text_x, text_y), text, fill=(0, 0, 0, 180), font=font)

            # Apply slight blur and composite
            blurred_text_layer = text_layer.filter(ImageFilter.GaussianBlur(radius=1.0))
            pil_image = Image.alpha_composite(pil_image, blurred_text_layer)

        # Save and return output image path
        output_image_path = "/tmp/output_image.png"
        pil_image.convert("RGB").save(output_image_path)
        return output_image_path

    except Exception as e:
        logging.error(f"Error during image processing: {e}")
        return None

# Gradio interface function
def gradio_inference(image, text):
    result_path = process_image(image, text)
    if result_path:
        return result_path, result_path, "Processing complete! Download the image below."
    return None, None, "An error occurred while processing the image. Please check the logs."

# Gradio interface
interface = gr.Interface(
    fn=gradio_inference,
    inputs=[
        gr.Image(type="pil", label="Upload an Image"),
        gr.Textbox(label="Enter Text to Overlay")
    ],
    outputs=[
        gr.Image(label="Processed Image Preview"),  # Preview processed image
        gr.File(label="Download Processed Image"),  # Download the image
        gr.Textbox(label="Status")  # Status message
    ],
    title="Roboflow Detection with Text Overlay",
    description="Upload an image, enter text to overlay, and let the Roboflow model process the image. Preview or download the output image below.",
    allow_flagging="never"
)

# Launch the Gradio app
if __name__ == "__main__":
    interface.launch(share=True)