Spaces:
No application file
No application file
File size: 3,526 Bytes
e0547b4 | 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 | from PIL import Image
import numpy as np
import gradio as gr
import os
import tempfile
def perform_ela_with_red_dots(input_image, quality=90, scale=10, threshold=50):
"""
Perform Error Level Analysis (ELA) on an image and overlay red dots on tampered areas.
Parameters:
- input_image: PIL Image object (input image)
- quality: JPEG quality for recompression (default: 90)
- scale: Factor to amplify differences for visibility (default: 10)
- threshold: Threshold for highlighting significant differences (default: 50)
Returns:
- result_image: PIL Image object with ELA red dots overlaid
"""
try:
# Step 1: Convert input image to RGB
original = input_image.convert('RGB')
original_array = np.array(original)
# Step 2: Save the image at a slightly lower quality (recompress)
with tempfile.NamedTemporaryFile(suffix=".jpg", delete=False) as temp_file:
temp_path = temp_file.name
original.save(temp_path, "JPEG", quality=quality)
# Step 3: Load the recompressed image
recompressed = Image.open(temp_path).convert('RGB')
recompressed_array = np.array(recompressed)
# Step 4: Compute the absolute difference between original and recompressed
diff = np.abs(original_array - recompressed_array)
# Step 5: Amplify the differences for better visibility
ela = diff * scale
ela = np.clip(ela, 0, 255).astype(np.uint8)
# Step 6: Create a red-dot overlay for significant differences
red_dots = np.zeros_like(original_array)
mask = ela > threshold # Highlight areas above the threshold
red_dots[mask] = [255, 0, 0] # Red color for significant differences
# Step 7: Blend the red dots with the original image
alpha = 0.5 # Transparency for blending
blended = original_array.copy()
blended = blended * (1 - alpha) + red_dots * alpha
blended = blended.astype(np.uint8)
# Step 8: Convert the result back to a PIL Image
result_image = Image.fromarray(blended)
# Step 9: Clean up the temporary file
if os.path.exists(temp_path):
os.remove(temp_path)
return result_image
except Exception as e:
raise Exception(f"Error performing ELA: {e}")
# Gradio interface function
def ela_interface(input_image):
"""
Gradio interface to process an image and return the ELA result.
Parameters:
- input_image: Input image uploaded via Gradio
Returns:
- result_image: Processed image with ELA red dots
"""
if input_image is None:
raise gr.Error("Please upload an image to process.")
# Perform ELA on the uploaded image
result_image = perform_ela_with_red_dots(input_image, quality=90, scale=10, threshold=50)
return result_image
# Define the Gradio interface
interface = gr.Interface(
fn=ela_interface, # Function to call
inputs=gr.Image(type="pil", label="Upload Image (JPEG)"), # Input: Image upload
outputs=gr.Image(type="pil", label="ELA Result with Red Dots"), # Output: Processed image
title="Error Level Analysis (ELA) for Image Tampering Detection",
description="Upload a JPEG image to perform Error Level Analysis (ELA). Areas with potential tampering will be highlighted with red dots.",
allow_flagging="never"
)
# Launch the Gradio app
if __name__ == "__main__":
interface.launch(server_name="0.0.0.0", server_port=7860) |