from PIL import Image, ImageEnhance, ImageFilter import numpy as np import torch import cv2 as cv def tensor2pil(tensor): """ PyTorchテンソルをPIL画像に変換する。 BHWC形式で値が[0, 1]の範囲のテンソルを想定。 """ # テンソルがCPUにあることを確認 tensor = tensor.cpu() # numpy配列に変換し、0-255の範囲に正規化 img_array = tensor.numpy() img_array = np.clip(img_array * 255.0, 0, 255).astype(np.uint8) # バッチ処理の場合は最初の画像を取得 if len(img_array.shape) == 4: img_array = img_array[0] # バッチ次元を削除 return Image.fromarray(img_array, mode='RGB') def pil2tensor(image, original_shape): """ PIL画像をPyTorchテンソルに変換し、元の形状に合わせる。 BHWC形状で値が[0, 1]の範囲のテンソルを返す。 """ # numpy配列に変換し、[0, 1]に正規化 img_array = np.array(image).astype(np.float32) / 255.0 # 元の形状に合わせてバッチ次元を追加 img_array = img_array[np.newaxis, ...] # PyTorchテンソルに変換 tensor = torch.from_numpy(img_array).float() return tensor def medianFilter(image, radius, num_samples, threshold): """ 画像に高品質なメディアンフィルタを適用する """ # PILからCV2形式に変換 cv_image = cv.cvtColor(np.array(image), cv.COLOR_RGB2BGR) # エッジを保持しながらスムージングを適用 blurred = cv.bilateralFilter(cv_image, radius, num_samples, threshold) # PIL形式に戻す return Image.fromarray(cv.cvtColor(blurred, cv.COLOR_BGR2RGB)) class FluxLightingAndColor: """ FluxLightingAndColor - Enhanced Version 画像の照明と色調を自動解析し調整するためのノードクラス 新機能: - 画像解析による自動パラメーター補完 - 色温度分析 - 輝度分布解析 - 彩度解析 - カスタマイズ可能な自動調整 """ @classmethod def INPUT_TYPES(s): """ 入力パラメータの定義 """ return { "required": { "image": ("IMAGE",), "auto_analysis": ("BOOLEAN", {"default": True}), "analysis_strength": ("FLOAT", {"default": 0.7, "min": 0.0, "max": 1.0, "step": 0.1}), "black_level": ("FLOAT", {"default": 0.0, "min": 0.0, "max": 1.0, "step": 0.01}), "mid_level": ("FLOAT", {"default": 0.5, "min": 0.0, "max": 1.0, "step": 0.01}), "white_level": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 1.0, "step": 0.01}), "red_level": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 2.0, "step": 0.01}), "green_level": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 2.0, "step": 0.01}), "blue_level": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 2.0, "step": 0.01}), "brightness": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 2.0, "step": 0.01}), "saturation": ("FLOAT", {"default": 1.0, "min": 0.0, "max": 2.0, "step": 0.01}), }, "optional": { "depth": ("IMAGE",), "dof_mode": (["none", "mock", "gaussian", "box"],), "dof_radius": ("INT", {"default": 8, "min": 1, "max": 128, "step": 1}), "dof_samples": ("INT", {"default": 1, "min": 1, "max": 3, "step": 1}), "auto_brightness": ("BOOLEAN", {"default": True}), "auto_color_balance": ("BOOLEAN", {"default": True}), "auto_saturation": ("BOOLEAN", {"default": True}), "auto_contrast": ("BOOLEAN", {"default": True}), "debug_mode": ("BOOLEAN", {"default": False}), } } RETURN_TYPES = ("IMAGE", "STRING") RETURN_NAMES = ("image", "analysis_report") FUNCTION = "apply_lighting_and_color" CATEGORY = "image/adjustments" def analyze_image_characteristics(self, img_array, debug=False): """ 画像の特性を解析し、最適な調整パラメーターを算出する Returns: - dict: 解析結果と推奨パラメーター """ analysis = {} # 輝度分析 luminance = 0.299 * img_array[:,:,0] + 0.587 * img_array[:,:,1] + 0.114 * img_array[:,:,2] # 基本統計 analysis['mean_luminance'] = float(np.mean(luminance)) analysis['std_luminance'] = float(np.std(luminance)) analysis['min_luminance'] = float(np.min(luminance)) analysis['max_luminance'] = float(np.max(luminance)) # ヒストグラム分析 hist, bins = np.histogram(luminance.flatten(), bins=256, range=(0, 1)) analysis['histogram'] = hist # 色温度分析（簡易版） mean_r = float(np.mean(img_array[:,:,0])) mean_g = float(np.mean(img_array[:,:,1])) mean_b = float(np.mean(img_array[:,:,2])) analysis['mean_rgb'] = [mean_r, mean_g, mean_b] # 色温度推定 (青が強い = 冷たい、赤が強い = 暖かい) if mean_b > mean_r: analysis['color_temperature'] = 'cool' analysis['temp_bias'] = mean_b - mean_r else: analysis['color_temperature'] = 'warm' analysis['temp_bias'] = mean_r - mean_b # 彩度分析 hsv_array = cv.cvtColor((img_array * 255).astype(np.uint8), cv.COLOR_RGB2HSV) saturation_values = hsv_array[:,:,1] / 255.0 analysis['mean_saturation'] = float(np.mean(saturation_values)) analysis['std_saturation'] = float(np.std(saturation_values)) # コントラスト分析 analysis['contrast_ratio'] = analysis['max_luminance'] - analysis['min_luminance'] # 推奨パラメーター計算 recommendations = self.calculate_recommendations(analysis, debug) analysis['recommendations'] = recommendations if debug: print("\n=== Image Analysis Results ===") print(f"Mean Luminance: {analysis['mean_luminance']:.3f}") print(f"Luminance Range: {analysis['min_luminance']:.3f} - {analysis['max_luminance']:.3f}") print(f"Color Temperature: {analysis['color_temperature']} (bias: {analysis['temp_bias']:.3f})") print(f"Mean Saturation: {analysis['mean_saturation']:.3f}") print(f"Contrast Ratio: {analysis['contrast_ratio']:.3f}") print("Recommendations:", recommendations) print("===============================\n") return analysis def calculate_recommendations(self, analysis, debug=False): """ 解析結果に基づいて推奨パラメーターを計算する """ recommendations = {} # 明るさ調整の推奨値 target_luminance = 0.5 current_luminance = analysis['mean_luminance'] if current_luminance < 0.3: # 暗い画像 recommendations['brightness'] = 1.2 + (0.3 - current_luminance) * 2 elif current_luminance > 0.7: # 明るい画像 recommendations['brightness'] = 0.8 + (0.7 - current_luminance) * 0.5 else: recommendations['brightness'] = 1.0 # コントラスト調整 if analysis['contrast_ratio'] < 0.5: # 低コントラスト recommendations['black_level'] = max(0, analysis['min_luminance'] - 0.05) recommendations['white_level'] = min(1, analysis['max_luminance'] + 0.05) else: recommendations['black_level'] = 0.0 recommendations['white_level'] = 1.0 recommendations['mid_level'] = 0.5 # 色温度補正 temp_bias = analysis['temp_bias'] if analysis['color_temperature'] == 'cool' and temp_bias > 0.05: # 冷たすぎる場合、赤を強化、青を減少 recommendations['red_level'] = 1.0 + min(temp_bias * 2, 0.3) recommendations['green_level'] = 1.0 recommendations['blue_level'] = 1.0 - min(temp_bias * 1.5, 0.2) elif analysis['color_temperature'] == 'warm' and temp_bias > 0.05: # 暖かすぎる場合、青を強化、赤を減少 recommendations['red_level'] = 1.0 - min(temp_bias * 1.5, 0.2) recommendations['green_level'] = 1.0 recommendations['blue_level'] = 1.0 + min(temp_bias * 2, 0.3) else: recommendations['red_level'] = 1.0 recommendations['green_level'] = 1.0 recommendations['blue_level'] = 1.0 # 彩度調整 mean_sat = analysis['mean_saturation'] if mean_sat < 0.4: # 彩度が低い recommendations['saturation'] = 1.0 + (0.4 - mean_sat) * 2 elif mean_sat > 0.8: # 彩度が高い recommendations['saturation'] = 1.0 - (mean_sat - 0.8) * 1.5 else: recommendations['saturation'] = 1.0 # 値を適切な範囲にクリップ recommendations['brightness'] = np.clip(recommendations['brightness'], 0.0, 2.0) recommendations['saturation'] = np.clip(recommendations['saturation'], 0.0, 2.0) recommendations['red_level'] = np.clip(recommendations['red_level'], 0.0, 2.0) recommendations['green_level'] = np.clip(recommendations['green_level'], 0.0, 2.0) recommendations['blue_level'] = np.clip(recommendations['blue_level'], 0.0, 2.0) return recommendations def blend_parameters(self, user_params, recommendations, strength, auto_flags, debug=False): """ ユーザー設定と推奨値をブレンドする Parameters: - user_params: ユーザーが設定したパラメーター - recommendations: 解析による推奨パラメーター - strength: 自動調整の強度 (0.0-1.0) - auto_flags: 各調整項目の自動有効フラグ """ blended = {} param_mapping = { 'brightness': 'auto_brightness', 'saturation': 'auto_saturation', 'red_level': 'auto_color_balance', 'green_level': 'auto_color_balance', 'blue_level': 'auto_color_balance', 'black_level': 'auto_contrast', 'white_level': 'auto_contrast', 'mid_level': 'auto_contrast' } for param_name in ['brightness', 'saturation', 'red_level', 'green_level', 'blue_level', 'black_level', 'white_level', 'mid_level']: user_value = user_params.get(param_name, 1.0) recommended_value = recommendations.get(param_name, user_value) auto_flag_name = param_mapping.get(param_name, 'auto_brightness') if auto_flags.get(auto_flag_name, True): # 自動調整が有効な場合、ユーザー値と推奨値をブレンド blended[param_name] = user_value * (1 - strength) + recommended_value * strength else: # 自動調整が無効な場合、ユーザー値をそのまま使用 blended[param_name] = user_value if debug: print("\n=== Parameter Blending ===") for param_name in blended: user_val = user_params.get(param_name, 1.0) rec_val = recommendations.get(param_name, user_val) final_val = blended[param_name] auto_flag = param_mapping.get(param_name, 'auto_brightness') enabled = auto_flags.get(auto_flag, True) print(f"{param_name}: User={user_val:.3f}, Rec={rec_val:.3f}, Final={final_val:.3f} (Auto: {enabled})") print("=========================\n") return blended def generate_analysis_report(self, analysis, blended_params, user_params): """ 解析結果のレポートを生成する """ report = [] report.append("=== Image Analysis Report ===") report.append(f"Mean Luminance: {analysis['mean_luminance']:.3f}") report.append(f"Luminance Range: {analysis['min_luminance']:.3f} - {analysis['max_luminance']:.3f}") report.append(f"Color Temperature: {analysis['color_temperature']} (bias: {analysis['temp_bias']:.3f})") report.append(f"Mean Saturation: {analysis['mean_saturation']:.3f}") report.append(f"Contrast Ratio: {analysis['contrast_ratio']:.3f}") report.append("") report.append("=== Applied Adjustments ===") for param_name in ['brightness', 'saturation', 'red_level', 'green_level', 'blue_level']: user_val = user_params.get(param_name, 1.0) final_val = blended_params.get(param_name, user_val) change = final_val - user_val if abs(change) > 0.01: sign = "+" if change > 0 else "" report.append(f"{param_name}: {user_val:.3f} → {final_val:.3f} ({sign}{change:.3f})") else: report.append(f"{param_name}: {final_val:.3f} (no change)") report.append("=============================") return "\n".join(report) def apply_dof(self, img, depth_map=None, mode='none', radius=8, samples=1, debug=False): """ 被写界深度(DoF)エフェクトを適用する """ if mode == 'none' or depth_map is None: if debug: print("DoF: Skipped (mode: none or no depth map)") return img if debug: print(f"DoF: Applying {mode} blur with radius {radius} and {samples} samples") # Resize depth map to match image size and convert to grayscale depth_map = depth_map.resize(img.size).convert('L') # Apply blur based on selected mode if mode == 'mock': blurred = medianFilter(img, radius, (radius * 1500), 75) elif mode == 'gaussian': blurred = img.filter(ImageFilter.GaussianBlur(radius=radius)) elif mode == 'box': blurred = img.filter(ImageFilter.BoxBlur(radius)) else: return img blurred = blurred.convert(img.mode) # Apply multiple samples if requested if samples > 1: result = None for i in range(samples): if not result: result = Image.composite(img, blurred, depth_map) else: result = Image.composite(result, blurred, depth_map) if debug: print(f"DoF: Applied sample {i+1}/{samples}") else: result = Image.composite(img, blurred, depth_map).convert('RGB') return result def adjust_levels(self, img_array, black_level, mid_level, white_level, debug=False): """ 画像のレベル調整を行う """ if debug: print(f"Levels: Adjusting (black: {black_level:.3f}, mid: {mid_level:.3f}, white: {white_level:.3f})") # Apply level adjustments img_array = (img_array - black_level) / (white_level - black_level) img_array = np.clip(img_array, 0, 1) # Apply gamma correction based on mid level if mid_level != 0.5: gamma = np.log(0.5) / np.log(mid_level + 1e-8) # avoid division by zero img_array = np.power(img_array, gamma) return img_array def adjust_channels(self, img_array, red_level, green_level, blue_level, debug=False): """ RGB各チャンネルの強度を調整する """ if debug: print(f"Channels: Adjusting (R: {red_level:.3f}, G: {green_level:.3f}, B: {blue_level:.3f})") # Split and adjust each channel img_array[:,:,0] = np.clip(img_array[:,:,0] * red_level, 0, 1) img_array[:,:,1] = np.clip(img_array[:,:,1] * green_level, 0, 1) img_array[:,:,2] = np.clip(img_array[:,:,2] * blue_level, 0, 1) return img_array def apply_lighting_and_color(self, image, auto_analysis, analysis_strength, black_level, mid_level, white_level, red_level, green_level, blue_level, brightness, saturation, depth=None, dof_mode="none", dof_radius=8, dof_samples=1, auto_brightness=True, auto_color_balance=True, auto_saturation=True, auto_contrast=True, debug_mode=False): """ メインの処理関数。画像解析と自動調整機能を含む。 """ try: if debug_mode: print("\n=== FluxLightingAndColor Enhanced Starting ===") print(f"Input tensor shape: {image.shape}, dtype: {image.dtype}") print(f"Auto analysis: {auto_analysis}, Strength: {analysis_strength}") # 1. 入力テンソルをPIL画像に変換 img_pil = tensor2pil(image) if debug_mode: print("Step 1: Converted input tensor to PIL image") # 2. 画像解析（自動調整が有効な場合） analysis_report = "No analysis performed" final_params = { 'black_level': black_level, 'mid_level': mid_level, 'white_level': white_level, 'red_level': red_level, 'green_level': green_level, 'blue_level': blue_level, 'brightness': brightness, 'saturation': saturation } if auto_analysis: img_array_for_analysis = np.array(img_pil).astype(float) / 255.0 analysis = self.analyze_image_characteristics(img_array_for_analysis, debug_mode) user_params = { 'black_level': black_level, 'mid_level': mid_level, 'white_level': white_level, 'red_level': red_level, 'green_level': green_level, 'blue_level': blue_level, 'brightness': brightness, 'saturation': saturation } auto_flags = { 'auto_brightness': auto_brightness, 'auto_color_balance': auto_color_balance, 'auto_saturation': auto_saturation, 'auto_contrast': auto_contrast } final_params = self.blend_parameters( user_params, analysis['recommendations'], analysis_strength, auto_flags, debug_mode ) analysis_report = self.generate_analysis_report(analysis, final_params, user_params) if debug_mode: print("Step 2: Completed image analysis and parameter blending") # 3. 深度マップがある場合はDoFを適用 if depth is not None and dof_mode != "none": depth_pil = tensor2pil(depth) img_pil = self.apply_dof(img_pil, depth_pil, dof_mode, dof_radius, dof_samples, debug_mode) if debug_mode: print("Step 3: Applied depth of field effect") # 4. 処理用にnumpy配列に変換 img_array = np.array(img_pil).astype(float) / 255.0 # 5. 明るさ調整を適用 if debug_mode: print(f"Step 5: Applying brightness: {final_params['brightness']:.3f}") img_array = np.power(img_array, 0.7) * final_params['brightness'] # 6. レベル調整を適用 img_array = self.adjust_levels( img_array, final_params['black_level'], final_params['mid_level'], final_params['white_level'], debug_mode ) # 7. チャンネル調整を適用 img_array = self.adjust_channels( img_array, final_params['red_level'], final_params['green_level'], final_params['blue_level'], debug_mode ) # 8. エンハンス処理用にPIL画像に再変換 processed = Image.fromarray((np.clip(img_array * 255.0, 0, 255)).astype(np.uint8)) # 9. 彩度を適用 if debug_mode: print(f"Step 9: Applying saturation: {final_params['saturation']:.3f}") enhancer = ImageEnhance.Color(processed) processed = enhancer.enhance(final_params['saturation']) # 10. 最終的なコントラストを適用 contrast_boost = 1.3 if auto_analysis and auto_contrast: # 既にコントラストが高い場合は控えめに if analysis['contrast_ratio'] > 0.7: contrast_boost = 1.1 if debug_mode: print(f"Step 10: Applying final contrast boost: {contrast_boost}") enhancer = ImageEnhance.Contrast(processed) processed = enhancer.enhance(contrast_boost) # 11. テンソルに再変換 result = pil2tensor(processed, image.shape) if debug_mode: print(f"Output tensor shape: {result.shape}, dtype: {result.dtype}") print("=== Enhanced Processing Complete ===\n") return (result, analysis_report) except Exception as e: print(f"Error in apply_lighting_and_color: {str(e)}") import traceback traceback.print_exc() return (image, f"Error occurred: {str(e)}") # Node registration NODE_CLASS_MAPPINGS = { "FluxLightingAndColor": FluxLightingAndColor } NODE_DISPLAY_NAME_MAPPINGS = { "FluxLightingAndColor": "Flux Lighting & Color (Enhanced)" }