Image Understanding

# Using Python PIL (Pillow) to prepare an image for AI analysis
# This shows image processing before sending to a multimodal AI API

from PIL import Image
import base64
import io
import os

def prepare_image_for_ai(image_path, max_size=(1024, 1024)):
    """
    Resize and convert an image to base64 for sending to an AI API.
    Most AI image APIs accept base64-encoded images.

    Parameters:
    - image_path: path to the local image file
    - max_size: maximum dimensions to resize the image to (reduces cost)
    """
    # Install: pip install Pillow

    with Image.open(image_path) as img:
        # Get original dimensions
        original_size = img.size
        print(f"Original image size: {original_size[0]}x{original_size[1]} pixels")

        # Resize if larger than max_size (keeps aspect ratio)
        img.thumbnail(max_size, Image.LANCZOS)
        resized_size = img.size
        print(f"Resized to: {resized_size[0]}x{resized_size[1]} pixels")

        # Convert to RGB if needed (some images are RGBA or have other modes)
        if img.mode != 'RGB':
            img = img.convert('RGB')

        # Convert to base64
        buffer = io.BytesIO()
        img.save(buffer, format='JPEG', quality=85)
        image_bytes = buffer.getvalue()
        base64_image = base64.b64encode(image_bytes).decode('utf-8')

        print(f"Base64 string length: {len(base64_image)} characters")
        print(f"Ready to send to AI API: data:image/jpeg;base64,{base64_image[:30]}...")

    return base64_image

# Usage example (requires a real image file)
# base64_img = prepare_image_for_ai("photo.jpg")

print("Image preparation utility ready.")
print("Install Pillow: pip install Pillow")
print("Then call: prepare_image_for_ai('your_image.jpg')")