Update app.py

app.py

@@ -36,67 +36,33 @@ tokenizer = AutoTokenizer.from_pretrained(
 )
 
 
-def find_closest_aspect_ratio(aspect_ratio, target_ratios, width, height, image_size):
-    best_ratio_diff = float('inf')
-    best_ratio = (1, 1)
-    area = width * height
-    for ratio in target_ratios:
-        target_aspect_ratio = ratio[0] / ratio[1]
-        ratio_diff = abs(aspect_ratio - target_aspect_ratio)
-        if ratio_diff < best_ratio_diff:
-            best_ratio_diff = ratio_diff
-            best_ratio = ratio
-        elif ratio_diff == best_ratio_diff:
-            if area > 0.5 * image_size * image_size * ratio[0] * ratio[1]:
-                best_ratio = ratio
-    return best_ratio
-
-def dynamic_preprocess(image, min_num=1, max_num=12, image_size=448, use_thumbnail=False):
-    # Ensure the image is a PIL Image
-    if not isinstance(image, Image.Image):
-        image = Image.fromarray(image)
-        
+def preprocessing(image, image_size=448):
+    """
+    Apply enhancement filters and pad the image to match the target size while keeping full content.
+    """
+    # Convert input to a PIL Image (if it isn’t already)
+    image = Image.fromarray(np.array(image))
+    
+    # Apply enhancement filters
+    image = ImageEnhance.Sharpness(image).enhance(2.0)   # Increase sharpness
+    image = ImageEnhance.Contrast(image).enhance(1.5)    # Increase contrast
+    image = ImageEnhance.Brightness(image).enhance(0.8)  # Reduce brightness
+
+    # Get original dimensions
     orig_width, orig_height = image.size
-    aspect_ratio = orig_width / orig_height
-
-    # Calculate the existing image aspect ratio
-    target_ratios = set(
-        (i, j) for n in range(min_num, max_num + 1) 
-        for i in range(1, n + 1) 
-        for j in range(1, n + 1) 
-        if i * j <= max_num and i * j >= min_num
-    )
-    target_ratios = sorted(target_ratios, key=lambda x: x[0] * x[1])
-
-    # Find the closest aspect ratio to the target
-    target_aspect_ratio = find_closest_aspect_ratio(
-        aspect_ratio, target_ratios, orig_width, orig_height, image_size
-    )
-
-    # Calculate the target width and height
-    target_width = image_size * target_aspect_ratio[0]
-    target_height = image_size * target_aspect_ratio[1]
-    blocks = target_aspect_ratio[0] * target_aspect_ratio[1]
-
-    # Resize the image
-    resized_img = image.resize((target_width, target_height))
-    processed_images = []
-    for i in range(blocks):
-        # Calculate the crop box for each block
-        box = (
-            (i % (target_width // image_size)) * image_size,
-            (i // (target_width // image_size)) * image_size,
-            ((i % (target_width // image_size)) + 1) * image_size,
-            ((i // (target_width // image_size)) + 1) * image_size
-        )
-        # Split the image
-        split_img = resized_img.crop(box)
-        processed_images.append(split_img)
-    assert len(processed_images) == blocks
-    if use_thumbnail and len(processed_images) != 1:
-        thumbnail_img = image.resize((image_size, image_size))
-        processed_images.append(thumbnail_img)
-    return processed_images[0]
+
+    # Determine the padding needed to fit the image within a square of size `image_size`
+    pad_x = max(image_size - orig_width, 0)
+    pad_y = max(image_size - orig_height, 0)
+
+    # Create a new blank image with a white background
+    padded_image = Image.new("RGB", (orig_width + pad_x, orig_height + pad_y), (255, 255, 255))
+
+    # Paste the original image in the center
+    padded_image.paste(image, (pad_x // 2, pad_y // 2))
+
+    return padded_image
+