File size: 10,936 Bytes
338d95d |
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 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 |
"""
Image processing utilities for CompI Phase 2.E: Style Reference/Example Image Integration
This module provides utilities for:
- Image loading from files and URLs
- Image validation and preprocessing
- Style analysis and feature extraction
- Image format conversion and optimization
"""
import os
import io
import requests
import hashlib
from typing import Optional, Tuple, Dict, Any, Union, List
from pathlib import Path
import logging
import torch
import numpy as np
from PIL import Image, ImageStat, ImageFilter
import cv2
from src.utils.logging_utils import setup_logger
logger = setup_logger(__name__)
class ImageProcessor:
"""
Handles image loading, validation, and preprocessing for style reference
"""
def __init__(self, max_size: Tuple[int, int] = (1024, 1024)):
self.max_size = max_size
self.supported_formats = {'.jpg', '.jpeg', '.png', '.bmp', '.tiff', '.webp'}
def load_image_from_url(
self,
url: str,
timeout: int = 10,
max_file_size: int = 10 * 1024 * 1024 # 10MB
) -> Optional[Image.Image]:
"""
Load image from URL with validation and error handling
Args:
url: Image URL
timeout: Request timeout in seconds
max_file_size: Maximum file size in bytes
Returns:
PIL Image or None if failed
"""
try:
logger.info(f"Loading image from URL: {url}")
# Validate URL format
if not url.startswith(('http://', 'https://')):
logger.error(f"Invalid URL format: {url}")
return None
# Make request with headers to avoid blocking
headers = {
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36'
}
response = requests.get(url, timeout=timeout, headers=headers, stream=True)
response.raise_for_status()
# Check content type
content_type = response.headers.get('content-type', '').lower()
if not any(img_type in content_type for img_type in ['image/', 'jpeg', 'png', 'webp']):
logger.error(f"Invalid content type: {content_type}")
return None
# Check file size
content_length = response.headers.get('content-length')
if content_length and int(content_length) > max_file_size:
logger.error(f"File too large: {content_length} bytes")
return None
# Load image data
image_data = io.BytesIO()
downloaded_size = 0
for chunk in response.iter_content(chunk_size=8192):
downloaded_size += len(chunk)
if downloaded_size > max_file_size:
logger.error(f"File too large during download: {downloaded_size} bytes")
return None
image_data.write(chunk)
image_data.seek(0)
# Open and validate image
image = Image.open(image_data)
image = image.convert('RGB')
logger.info(f"Successfully loaded image: {image.size}")
return image
except requests.exceptions.RequestException as e:
logger.error(f"Request error loading image from {url}: {e}")
return None
except Exception as e:
logger.error(f"Error loading image from {url}: {e}")
return None
def load_image_from_file(self, file_path: Union[str, Path]) -> Optional[Image.Image]:
"""
Load image from local file with validation
Args:
file_path: Path to image file
Returns:
PIL Image or None if failed
"""
try:
file_path = Path(file_path)
if not file_path.exists():
logger.error(f"File does not exist: {file_path}")
return None
if file_path.suffix.lower() not in self.supported_formats:
logger.error(f"Unsupported format: {file_path.suffix}")
return None
image = Image.open(file_path)
image = image.convert('RGB')
logger.info(f"Successfully loaded image from file: {image.size}")
return image
except Exception as e:
logger.error(f"Error loading image from {file_path}: {e}")
return None
def preprocess_image(
self,
image: Image.Image,
target_size: Optional[Tuple[int, int]] = None,
maintain_aspect_ratio: bool = True
) -> Image.Image:
"""
Preprocess image for stable diffusion
Args:
image: Input PIL Image
target_size: Target size (width, height)
maintain_aspect_ratio: Whether to maintain aspect ratio
Returns:
Preprocessed PIL Image
"""
if target_size is None:
target_size = (512, 512) # Default SD size
try:
# Resize image
if maintain_aspect_ratio:
image.thumbnail(target_size, Image.Resampling.LANCZOS)
# Create new image with target size and paste resized image
new_image = Image.new('RGB', target_size, (255, 255, 255))
paste_x = (target_size[0] - image.width) // 2
paste_y = (target_size[1] - image.height) // 2
new_image.paste(image, (paste_x, paste_y))
image = new_image
else:
image = image.resize(target_size, Image.Resampling.LANCZOS)
logger.info(f"Preprocessed image to size: {image.size}")
return image
except Exception as e:
logger.error(f"Error preprocessing image: {e}")
return image
def analyze_image_properties(self, image: Image.Image) -> Dict[str, Any]:
"""
Analyze image properties for style reference
Args:
image: PIL Image to analyze
Returns:
Dictionary of image properties
"""
try:
# Basic properties
width, height = image.size
aspect_ratio = width / height
# Color analysis
stat = ImageStat.Stat(image)
avg_brightness = sum(stat.mean) / len(stat.mean)
avg_contrast = sum(stat.stddev) / len(stat.stddev)
# Convert to numpy for additional analysis
img_array = np.array(image)
# Color distribution
r_mean, g_mean, b_mean = np.mean(img_array, axis=(0, 1))
color_variance = np.var(img_array, axis=(0, 1))
# Edge detection for complexity
gray = cv2.cvtColor(img_array, cv2.COLOR_RGB2GRAY)
edges = cv2.Canny(gray, 50, 150)
edge_density = np.sum(edges > 0) / (width * height)
properties = {
'dimensions': (width, height),
'aspect_ratio': aspect_ratio,
'brightness': avg_brightness,
'contrast': avg_contrast,
'color_means': (float(r_mean), float(g_mean), float(b_mean)),
'color_variance': color_variance.tolist(),
'edge_density': float(edge_density),
'file_size_pixels': width * height
}
logger.info(f"Analyzed image properties: {properties}")
return properties
except Exception as e:
logger.error(f"Error analyzing image properties: {e}")
return {}
def generate_image_hash(self, image: Image.Image) -> str:
"""
Generate hash for image deduplication
Args:
image: PIL Image
Returns:
MD5 hash string
"""
try:
# Convert image to bytes
img_bytes = io.BytesIO()
image.save(img_bytes, format='PNG')
img_bytes = img_bytes.getvalue()
# Generate hash
hash_md5 = hashlib.md5(img_bytes)
return hash_md5.hexdigest()
except Exception as e:
logger.error(f"Error generating image hash: {e}")
return ""
class StyleAnalyzer:
"""
Analyzes style characteristics of reference images
"""
def __init__(self):
self.style_keywords = {
'realistic': ['photo', 'realistic', 'detailed', 'sharp'],
'artistic': ['painting', 'artistic', 'brushstrokes', 'canvas'],
'anime': ['anime', 'manga', 'cartoon', 'stylized'],
'abstract': ['abstract', 'geometric', 'surreal', 'conceptual'],
'vintage': ['vintage', 'retro', 'aged', 'classic'],
'modern': ['modern', 'contemporary', 'clean', 'minimal']
}
def suggest_style_keywords(self, image_properties: Dict[str, Any]) -> List[str]:
"""
Suggest style keywords based on image analysis
Args:
image_properties: Properties from analyze_image_properties
Returns:
List of suggested style keywords
"""
suggestions = []
try:
brightness = image_properties.get('brightness', 128)
contrast = image_properties.get('contrast', 50)
edge_density = image_properties.get('edge_density', 0.1)
# Brightness-based suggestions
if brightness < 100:
suggestions.extend(['dark', 'moody', 'dramatic'])
elif brightness > 180:
suggestions.extend(['bright', 'light', 'airy'])
# Contrast-based suggestions
if contrast > 80:
suggestions.extend(['high contrast', 'bold', 'striking'])
elif contrast < 30:
suggestions.extend(['soft', 'gentle', 'muted'])
# Edge density-based suggestions
if edge_density > 0.2:
suggestions.extend(['detailed', 'complex', 'intricate'])
elif edge_density < 0.05:
suggestions.extend(['smooth', 'simple', 'minimalist'])
return list(set(suggestions)) # Remove duplicates
except Exception as e:
logger.error(f"Error suggesting style keywords: {e}")
return []
|