import os import re import base64 import warnings import sys # Add sys import for debugging import ctypes # For short path conversion from ctypes import wintypes # For short path conversion # import opencc # <-- opencc 导入会保留，但路径处理提前 # --- Manually add D:\Lib\site-packages to sys.path --- site_packages_path = r'D:\Lib\site-packages' # Corrected path separator if site_packages_path not in sys.path: sys.path.append(site_packages_path) # --- End of manual addition --- # --- Manually add path for opencc if needed --- opencc_site_packages_path = r'C:\Users\86138\AppData\Roaming\Python\Python311\site-packages' # Corrected path separator if opencc_site_packages_path not in sys.path: sys.path.insert(0, opencc_site_packages_path) # Insert at the beginning print(f"--- Debug --- Added to sys.path for opencc: {opencc_site_packages_path}") # --- End of manual addition for opencc --- # Now try importing opencc import opencc print(f"--- Debug --- Attempting to import imageio in: {__file__}") # Debug print print(f"--- Debug --- Python executable: {sys.executable}") # Debug print print(f"--- Debug --- sys.path AFTER manual add: {sys.path}") # Debug print, note the change in message try: # Debug block import imageio as test_imageio_module print(f"--- Debug --- Found 'imageio' at: {test_imageio_module.__file__}") print(f"--- Debug --- Version of 'imageio': {test_imageio_module.__version__}") except ImportError as e: print(f"--- Debug --- ImportError for imageio: {e}") except AttributeError: # Handle cases where __file__ or __version__ might be missing print(f"--- Debug --- Found 'imageio', but cannot get __file__ or __version__.") # The original import line import imageio import whisper import numpy as np from PIL import Image from skimage.metrics import structural_similarity as ssim from collections import defaultdict import subprocess from jinja2 import Environment import cv2 from scipy.signal import find_peaks from skimage.feature import hog from skimage.color import rgb2gray import concurrent.futures import threading import queue import time import gc from functools import lru_cache import multiprocessing import signal import traceback import logging import json import shutil import importlib # 导入补丁模块 - 用于解决wkhtmltopdf依赖问题 try: import pdfkit_patch as pdfkit logging.info("已加载pdfkit补丁模块") except ImportError: logging.info("未找到pdfkit补丁模块，PDF生成功能可能不可用") # 设置环境变量，使用 OpenBLAS os.environ['OPENBLAS_NUM_THREADS'] = '1' os.environ['MKL_NUM_THREADS'] = '1' os.environ['NUMEXPR_NUM_THREADS'] = '1' os.environ['OMP_NUM_THREADS'] = '1' # FFmpeg路径配置 FFMPEG_BIN = os.path.join(os.path.dirname(os.path.abspath(__file__)), "ffmpeg", "bin") print(f"--- Debug --- FFMPEG_BIN calculated as: {repr(FFMPEG_BIN)}") # DEBUG LINE ADDED if not os.path.exists(FFMPEG_BIN): FFMPEG_BIN = "" # 如果目录不存在，使用系统环境变量中的FFmpeg print(f"--- Debug --- FFMPEG_BIN reset to empty string because path does not exist.") # DEBUG LINE ADDED # 配置日志 logging.basicConfig( level=logging.DEBUG, format='%(asctime)s - %(levelname)s - %(message)s', handlers=[ logging.FileHandler('process.log', encoding='utf-8'), logging.StreamHandler() ] ) # Helper function for short path def get_short_path_name(long_name): """Gets the short path name of a given long path.""" output_buf_size = 0 # First, call GetShortPathNameW with a null buffer to get the required buffer size. # We expect this to fail and return the size. output_buf_size = ctypes.windll.kernel32.GetShortPathNameW(long_name, None, 0) if output_buf_size == 0: # An error occurred, perhaps the path doesn't exist or another issue. print(f"--- Debug --- GetShortPathNameW failed to get buffer size for: {long_name}, error code: {ctypes.GetLastError()}") return long_name # Return original name if conversion fails output_buf = ctypes.create_unicode_buffer(output_buf_size) needed = ctypes.windll.kernel32.GetShortPathNameW(long_name, output_buf, output_buf_size) if needed == 0: print(f"--- Debug --- GetShortPathNameW failed to convert: {long_name}, error code: {ctypes.GetLastError()}") return long_name # Return original name if conversion fails else: return output_buf.value def check_dependencies(): try: # 检查FFmpeg try: subprocess.run(['ffmpeg', '-version'], capture_output=True, check=True) logging.info("FFmpeg 检查通过") except Exception as e: logging.error(f"FFmpeg 检查失败: {str(e)}") return False # 检查OpenCV try: import cv2 logging.info("OpenCV 检查通过") except Exception as e: logging.error(f"OpenCV 检查失败: {str(e)}") return False # 检查Whisper try: import whisper logging.info("Whisper 检查通过") except Exception as e: logging.error(f"Whisper 检查失败: {str(e)}") return False # 注意: wkhtmltopdf检查已禁用 # 使用pdfkit_patch模块解决wkhtmltopdf依赖问题 logging.info("wkhtmltopdf检查已禁用，仅生成HTML报告") logging.info("所有依赖项检查通过") return True except Exception as e: logging.error(f"依赖项检查失败: {str(e)}") return False # ======================== 全局配置 ======================== warnings.filterwarnings("ignore", message="FP16 is not supported on CPU; using FP32 instead") # 使用相对路径 BASE_DIR = os.path.dirname(os.path.abspath(__file__)) MODEL_DIR = os.path.join(BASE_DIR, "models") OUTPUT_DIR = os.path.join(BASE_DIR, "output") # 创建必要的目录 os.makedirs(MODEL_DIR, exist_ok=True) os.makedirs(OUTPUT_DIR, exist_ok=True) # 其他配置保持不变 SSIM_THRESHOLD = 0.85 # 关键帧去重阈值 FRAME_INTERVAL = 2 # 抽帧间隔（秒） TRANSITION_WORDS = ["接下来", "下一页", "如图"] # 过渡词过滤列 HOG_THRESHOLD = 0.7 # HOG特征相似度阈值 COLOR_THRESHOLD = 0.8 # 颜色直方图相似度阈值 WHISPER_MODEL = "small" # Whisper模型大小 PROFESSIONAL_TERMS = { "人工智能": "AI", "机器学习": "ML", "深度学习": "DL", "神经网络": "NN", "卷积神经网络": "CNN", "循环神经网络": "RNN", "自然语言处理": "NLP", "计算机视觉": "CV", "大数据": "Big Data", "云计算": "Cloud Computing" } # 专业术语词典 # 性能优化配置 MAX_WORKERS = max(1, multiprocessing.cpu_count() - 1) # 并行处理的工作线程数 BATCH_SIZE = 15 # 增加批处理大小 CACHE_SIZE = 150 # 增加缓存大小 MEMORY_LIMIT = 0.8 # 内存使用限制（占总内存的比例） TIMEOUT_SECONDS = 200 # 减少超时时间以加速处理流程 PROGRESS_UPDATE_INTERVAL = 1 # 进度更新间隔（秒） MAX_KEYFRAMES = 30 # 最大关键帧数量限制，超过此数量将进行抽样 # ======================================================== # 进度跟踪类 class ProgressTracker: def __init__(self, total_steps, description="处理中"): self.total_steps = total_steps self.current_step = 0 self.description = description self.start_time = time.time() self.last_update_time = self.start_time self._lock = threading.Lock() def update(self, step=1, message=None): with self._lock: self.current_step += step current_time = time.time() # 控制更新频率 if current_time - self.last_update_time >= PROGRESS_UPDATE_INTERVAL: elapsed = current_time - self.start_time progress = (self.current_step / self.total_steps) * 100 if message: print( f"[进度] {self.description}: {progress:.1f}% ({self.current_step}/{self.total_steps}) - {message}") else: print(f"[进度] {self.description}: {progress:.1f}% ({self.current_step}/{self.total_steps})") self.last_update_time = current_time def complete(self, message="完成"): with self._lock: elapsed = time.time() - self.start_time print(f"[完成] {self.description}: 100% - {message} (耗时: {elapsed:.1f}秒)") # 超时处理类 class TimeoutHandler: def __init__(self, timeout_seconds=TIMEOUT_SECONDS): self.timeout_seconds = timeout_seconds self.timer = None self._lock = threading.Lock() def start(self, operation_name): with self._lock: if self.timer: self.timer.cancel() self.timer = threading.Timer(self.timeout_seconds, self._timeout_callback, args=[operation_name]) self.timer.start() print(f"[信息] 开始{operation_name}，超时时间: {self.timeout_seconds}秒") def stop(self): with self._lock: if self.timer: self.timer.cancel() self.timer = None def _timeout_callback(self, operation_name): print(f"[警告] {operation_name}操作超时，正在尝试恢复...") # 这里可以添加恢复逻辑 # ---------------------- 核心功能模块 ---------------------- class VideoProcessor: def __init__(self): # os.environ["PATH"] = FFMPEG_BIN + os.pathsep + os.environ["PATH"] # COMMENTED OUT/MODIFIED if FFMPEG_BIN: # Only set if FFMPEG_BIN is not empty ffmpeg_exe_path = os.path.join(FFMPEG_BIN, 'ffmpeg.exe') print(f"--- Debug VideoProcessor --- Setting IMAGEIO_FFMPEG_EXE to: {repr(ffmpeg_exe_path)}") # DEBUG LINE ADDED os.environ['IMAGEIO_FFMPEG_EXE'] = ffmpeg_exe_path else: print("--- Debug VideoProcessor --- FFMPEG_BIN is empty, not setting IMAGEIO_FFMPEG_EXE.") # DEBUG LINE ADDED self.frame_cache = {} self.feature_cache = {} self._lock = threading.Lock() self.timeout_handler = TimeoutHandler() @staticmethod def check_ffmpeg(): """验证FFmpeg可用性""" try: subprocess.run(["ffmpeg", "-version"], check=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE) print("[系统] FFmpeg验证成功") return True except Exception as e: print(f"[错误] FFmpeg验证失败: {str(e)}") return False @lru_cache(maxsize=CACHE_SIZE) def calculate_color_histogram(self, frame_key): """计算颜色直方图特征（带缓存）""" frame = self.frame_cache.get(frame_key) if frame is None: return None hist = cv2.calcHist([frame], [0, 1, 2], None, [8, 8, 8], [0, 256, 0, 256, 0, 256]) cv2.normalize(hist, hist) return hist.flatten() @lru_cache(maxsize=CACHE_SIZE) def calculate_hog_features(self, frame_key): """计算HOG特征（带缓存）""" frame = self.frame_cache.get(frame_key) if frame is None: return None gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY) features = hog(gray, orientations=8, pixels_per_cell=(16, 16), cells_per_block=(1, 1), visualize=False) return features @staticmethod def is_ppt_transition(frame1, frame2): """检测PPT页面切换""" # 转换为灰度图 gray1 = cv2.cvtColor(frame1, cv2.COLOR_BGR2GRAY) gray2 = cv2.cvtColor(frame2, cv2.COLOR_BGR2GRAY) # 计算边缘 edges1 = cv2.Canny(gray1, 100, 200) edges2 = cv2.Canny(gray2, 100, 200) # 计算边缘差异 diff = cv2.absdiff(edges1, edges2) return np.mean(diff) > 50 # 阈值可调整 @staticmethod def is_blank_frame(frame, threshold=30): """检测是否为无信息帧（纯黑屏或纯白屏）""" try: # 转换为灰度图 gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY) # 计算图像统计特征 mean = np.mean(gray) std_dev = np.std(gray) # 检查是否为纯黑或纯白 is_black = mean < 10 and std_dev < 5 is_white = mean > 245 and std_dev < 5 # 检查是否有足够的细节 has_detail = std_dev > threshold return is_black or is_white or not has_detail except Exception as e: print(f"[警告] 检查无信息帧时出错: {str(e)}") return True def process_frame_batch(self, frames_batch, start_idx): """处理一批帧""" results = [] for i, frame in enumerate(frames_batch): idx = start_idx + i frame_key = f"frame_{idx}" self.frame_cache[frame_key] = frame results.append((idx, frame)) return results def extract_keyframes(self, video_path: str) -> tuple: """提取去重关键帧及其时间戳（多特征融合，并行处理）""" cap = None try: self.timeout_handler.start("关键帧提取") logging.info(f"[Debug extract_keyframes] Original video_path type: {type(video_path)}, value: {repr(video_path)}") # --- OpenCV VideoCapture for reading frames --- cap = cv2.VideoCapture(video_path) if not cap.isOpened(): logging.error(f"OpenCV: Failed to open video file: {video_path}") raise ValueError(f"无法打开视频文件: {video_path}") # 获取视频元数据 fps = cap.get(cv2.CAP_PROP_FPS) total_frames = int(cap.get(cv2.CAP_PROP_FRAME_COUNT)) duration = total_frames / fps if fps > 0 else 0 if total_frames <= 0 or fps <= 0 or duration <= 0: error_message = f"视频元数据读取不完整或无效: fps={fps}, total_frames={total_frames}, duration={duration}。" logging.error(error_message) raise ValueError(error_message) print(f"[信息] 视频元数据: 总帧数={total_frames}, 时长={duration:.2f}秒, FPS={fps:.2f}") keyframes = [] timestamps = [] prev_frame_bgr = None prev_frame_gray = None frame_count = 0 progress = ProgressTracker(total_frames, "关键帧提取 (OpenCV)") # 根据视频长度调整采样间隔 is_short_video = duration < 30 if is_short_video: sample_interval_frames = max(int(fps * 0.5), 1) # 短视频每0.5秒采样一次 else: sample_interval_frames = max(int(fps * 1.0), 1) # 长视频每1秒采样一次 logging.info(f"使用采样间隔: {sample_interval_frames}帧 (约{sample_interval_frames/fps:.1f}秒/帧)") while cap.isOpened(): ret, frame_bgr = cap.read() if not ret: break frame_count += 1 progress.update(1) # 按采样间隔处理帧 if frame_count % sample_interval_frames != 0: continue # 检查是否为空白帧 if self.is_blank_frame(frame_bgr, simplified=True): continue # 转换为灰度图用于SSIM计算 frame_gray = cv2.cvtColor(frame_bgr, cv2.COLOR_BGR2GRAY) # 与前一帧比较 if prev_frame_gray is not None: # 计算SSIM ssim_value = ssim(prev_frame_gray, frame_gray) logging.debug(f"[Debug SSIM] Frame {frame_count}, SSIM: {ssim_value:.4f}") if ssim_value > 0.95: logging.info(f"[Debug SSIM] Frame {frame_count} is too similar to previous frame, skipping.") continue # 保存关键帧 frame_rgb = cv2.cvtColor(frame_bgr, cv2.COLOR_BGR2RGB) frame_pil = Image.fromarray(frame_rgb) keyframes.append(frame_pil) timestamps.append(frame_count / fps) # 更新前一帧 prev_frame_bgr = frame_bgr.copy() prev_frame_gray = frame_gray.copy() # 如果没有提取到任何关键帧，进行强制采样 if not keyframes: logging.warning("自动提取关键帧为0，进行强制均匀采样") cap.set(cv2.CAP_PROP_POS_FRAMES, 0) # 重置到视频开始 frame_count = 0 # 强制采样间隔 force_sample_interval = max(total_frames // 10, 1) # 至少提取10帧 while cap.isOpened(): ret, frame_bgr = cap.read() if not ret: break frame_count += 1 if frame_count % force_sample_interval == 0: frame_rgb = cv2.cvtColor(frame_bgr, cv2.COLOR_BGR2RGB) frame_pil = Image.fromarray(frame_rgb) keyframes.append(frame_pil) timestamps.append(frame_count / fps) if not keyframes: raise RuntimeError("关键帧提取失败：未能提取到任何关键帧") print(f"[图像] 关键帧提取完成 (OpenCV)，共{len(keyframes)}帧") self.timeout_handler.stop() progress.complete(f"提取了{len(keyframes)}个关键帧 (OpenCV)") return keyframes, duration except Exception as e: logging.error(f"[错误] 关键帧提取失败 (OpenCV流程): {str(e)}") logging.error(traceback.format_exc()) self.timeout_handler.stop() return [], 0.0 finally: if cap and cap.isOpened(): cap.release() logging.info("OpenCV VideoCapture released in finally block.") def _is_frame_different(self, frame1_bgr_np, frame2_bgr_np, simplified=False, threshold=0.8): """简化版本的帧差异检测. Expects BGR NumPy arrays from OpenCV.""" if simplified: try: gray1 = cv2.cvtColor(frame1_bgr_np, cv2.COLOR_BGR2GRAY) gray2 = cv2.cvtColor(frame2_bgr_np, cv2.COLOR_BGR2GRAY) diff = cv2.absdiff(gray1, gray2) mean_diff = np.mean(diff) # 降低阈值，使检测更敏感 required_mean_diff = threshold * 3.0 # 从7.5降低到3.0 logging.debug(f"[Debug Diff] mean_diff: {mean_diff:.2f}, required_mean_diff (threshold={threshold:.2f}*3.0): {required_mean_diff:.2f}") return mean_diff > required_mean_diff except Exception as e_diff_internal: logging.error(f"Error in _is_frame_different simplified: {e_diff_internal}") return True # On error, assume different to avoid losing a frame else: # 完整的特征比较逻辑 (当前未被调用，因为 simplified=True) logging.warning("_is_frame_different called with simplified=False, but non-simplified path is not fully implemented.") return True def is_blank_frame(self, frame_bgr_np, simplified=False, threshold=20): """检测是否为无信息帧（支持简化版本）. Expects BGR NumPy array from OpenCV.""" try: gray = cv2.cvtColor(frame_bgr_np, cv2.COLOR_BGR2GRAY) mean = np.mean(gray) std_dev = np.std(gray) if simplified: # Simplified version for main loop # 调整阈值，使其更宽松 is_black = mean < 40 and std_dev < 20 # 从35/15调整到40/20 is_white = mean > 215 and std_dev < 20 # 从220/15调整到215/20 # 降低细节检测阈值 has_enough_detail = std_dev >= threshold * 0.8 # 降低阈值要求 is_actually_blank = (is_black or is_white) or not has_enough_detail logging.debug(f"[Debug BlankS] mean: {mean:.2f}, std_dev: {std_dev:.2f}, threshold: {threshold}, is_black: {is_black}, is_white: {is_white}, has_enough_detail: {has_enough_detail}, result: {is_actually_blank}") return is_actually_blank else: # Original more complex logic (currently not used by main path) is_black_orig = mean < 10 and std_dev < 5 is_white_orig = mean > 245 and std_dev < 5 has_detail_orig = std_dev > threshold is_actually_blank_orig = (is_black_orig or is_white_orig) or not has_detail_orig logging.debug(f"[Debug BlankNS] mean: {mean:.2f}, std_dev: {std_dev:.2f}, threshold: {threshold}, is_black: {is_black_orig}, is_white: {is_white_orig}, has_detail: {has_detail_orig}, result: {is_actually_blank_orig}") return is_actually_blank_orig except Exception as e_blank_internal: print(f"[警告] 检查无信息帧时出错 (OpenCV): {str(e_blank_internal)}") logging.error(f"Error in is_blank_frame: {e_blank_internal}") return False # On error, assume NOT blank to avoid wrongly discarding a frame @staticmethod def transcribe_audio(video_path: str, model_name: str = WHISPER_MODEL) -> list: """语音识别与时间戳获取（支持中英文混合，通过语言自动检测，并转换为简体中文）""" try: # 创建进度跟踪器 progress = ProgressTracker(100, "语音识别") progress.update(10, "加载模型") # 使用更大的模型提高准确率 model = whisper.load_model(model_name, device="cpu", download_root=MODEL_DIR) progress.update(20, "开始转写") logging.info(f"[Whisper] Starting transcription for: {video_path} with model: {model_name}. Language auto-detection ON.") result = model.transcribe( video_path, fp16=False, task="transcribe", verbose=True, initial_prompt=None ) detected_language = result.get("language", "unknown") logging.info(f"[Whisper] Transcription complete. Detected language: {detected_language}") progress.update(60, f"处理转写结果 (语言: {detected_language})") segments = result.get("segments", []) if detected_language == 'zh': logging.info("[Whisper] 检测到中文，将进行繁体到简体转换。") try: # 尝试不带 .json 后缀初始化，让库自行查找标准配置文件 converter = opencc.OpenCC('t2s') # <--- 修改点：移除了 .json for i, seg in enumerate(segments): original_text = seg['text'] simplified_text = converter.convert(original_text) if original_text != simplified_text: logging.debug(f"[OpenCC] 片段 {i} 转换: '{original_text[:30]}...' -> '{simplified_text[:30]}...'") seg['text'] = simplified_text logging.info("[OpenCC] 繁体到简体转换完成。") except Exception as e_opencc: logging.error(f"[OpenCC] 繁体到简体转换失败: {e_opencc}。将使用原始转录文本。") logging.info("[Whisper] 应用中文专业术语替换。") for i, seg in enumerate(segments): text = seg["text"] for cn, en in PROFESSIONAL_TERMS.items(): text = text.replace(cn, f"{cn}({en})") seg["text"] = text else: logging.info(f"[Whisper] Detected language is {detected_language}. Skipping Chinese char conversion and professional terms replacement.") if segments: progress.update(30, f"已处理 {len(segments)} 个片段的文本转换与术语替换 (如果适用)") else: progress.update(30, "无语音片段进行文本转换或术语替换处理") progress.complete(f"识别了{len(segments)}个语音片段") return segments except Exception as e: print(f"[错误] 语音识别失败: {str(e)}") return [] # ---------------------- 业务逻辑模块 ---------------------- class ContentAligner: @staticmethod def generate_page_intervals(timestamps: list, duration: float) -> list: """生成页面时间段""" intervals = [] for i in range(len(timestamps)): start = timestamps[i] end = timestamps[i + 1] if i < len(timestamps) - 1 else duration intervals.append((start, end)) return intervals @staticmethod @lru_cache(maxsize=CACHE_SIZE) def calculate_text_similarity(text1: str, text2: str) -> float: """计算文本相似度（带缓存）""" # 使用简单的词重叠度计算 words1 = set(re.findall(r'\w+', text1.lower())) words2 = set(re.findall(r'\w+', text2.lower())) if not words1 or not words2: return 0.0 intersection = words1.intersection(words2) union = words1.union(words2) return len(intersection) / len(union) @staticmethod def _process_segment(seg, seg_start, intervals, all_segments): """处理单个语音片段（用于并行处理）""" # 首先尝试时间戳匹配 for page_idx, (start, end) in enumerate(intervals): if start <= seg_start < end: return page_idx, seg # 如果时间戳匹配失败，尝试文本相似度匹配 best_page = None best_score = 0.0 for page_idx, (start, end) in enumerate(intervals): # 获取该页面的所有文本 page_text = " ".join([s["text"] for s in all_segments if start <= s["start"] < end]) similarity = ContentAligner.calculate_text_similarity(seg["text"], page_text) if similarity > best_score: best_score = similarity best_page = page_idx if best_page is not None: return best_page, seg return None @staticmethod def _filter_repetitive_segments(segments: list, min_repeats_for_deduplication: int = 3) -> list: """过滤连续重复的语音片段。如果一个片段的文本连续重复N次或更多，则只保留第一个实例。""" if not segments: return [] filtered_segments = [] i = 0 n = len(segments) while i < n: text_to_match = segments[i]['text'] # 计算当前文本连续重复的次数 count = 0 k = i while k < n and segments[k]['text'] == text_to_match: count += 1 k += 1 if count < min_repeats_for_deduplication: # 例如，重复1或2次，保留所有 filtered_segments.extend(segments[i : i + count]) else: # 例如，重复3次或更多，只保留第一个 filtered_segments.append(segments[i]) # 保留序列中的第一个片段 logging.info(f"文本去重：'{text_to_match[:50]}...' 连续出现 {count} 次，已保留1次。原始首片段信息：Start={segments[i]['start']}, End={segments[i]['end']}") i = k # 移动到下一个不同的文本块或列表末尾 return filtered_segments @staticmethod def find_best_match(segments: list, intervals: list) -> dict: """为每个语音片段找到最佳匹配的页面（并行处理）""" page_texts = defaultdict(list) unmatched_segments = [] # 创建进度跟踪器 progress = ProgressTracker(len(segments), "内容对齐") # 使用线程池进行并行处理 with concurrent.futures.ThreadPoolExecutor(max_workers=MAX_WORKERS) as executor: futures = [] for seg in segments: seg_start = seg["start"] future = executor.submit(ContentAligner._process_segment, seg, seg_start, intervals, segments) futures.append(future) # 收集结果 for i, future in enumerate(concurrent.futures.as_completed(futures)): try: result = future.result() if result: page_idx, seg = result page_texts[page_idx].append(seg) else: unmatched_segments.append(seg) progress.update(1, f"处理第{i + 1}/{len(segments)}个片段") except Exception as e: print(f"[警告] 处理语音片段时出错: {str(e)}") # 处理未匹配的片段 if unmatched_segments: print(f"[警告] 发现{len(unmatched_segments)}个未匹配的语音片段") # 将未匹配片段添加到最近的页面 for seg in unmatched_segments: closest_page = min(range(len(intervals)), key=lambda i: abs(seg["start"] - (intervals[i][0] + intervals[i][1]) / 2)) page_texts[closest_page].append(seg) progress.complete(f"对齐了{len(segments)}个语音片段") return page_texts @staticmethod def align_content(video_path: str, timestamps: list) -> list: """语音-画面对齐主逻辑（改进版，并行处理）""" timeout_handler_align = None # Initialize for finally block cap_align = None try: # 创建超时处理器 timeout_handler_align = TimeoutHandler() timeout_handler_align.start("内容对齐") # 获取视频时长 - Prefer OpenCV, then ffmpeg fallback (consistent with extract_keyframes) duration = 0 cap_align = cv2.VideoCapture(video_path) if cap_align.isOpened(): fps_align = cap_align.get(cv2.CAP_PROP_FPS) total_frames_align = int(cap_align.get(cv2.CAP_PROP_FRAME_COUNT)) if fps_align > 0 and total_frames_align > 0: duration = total_frames_align / fps_align logging.info(f"[AlignContent] OpenCV获取视频时长: {duration:.2f}秒") if duration <= 0 : # Fallback to ffmpeg CLI for duration logging.warning("[AlignContent] OpenCV未能获取有效时长，尝试FFmpeg CLI。") try: ffmpeg_exe_path = os.path.join(FFMPEG_BIN, 'ffmpeg.exe') if not (FFMPEG_BIN and os.path.exists(ffmpeg_exe_path)): ffmpeg_exe_path = 'ffmpeg' process = subprocess.run([ffmpeg_exe_path, '-i', video_path], capture_output=True, text=True, timeout=30) output_to_parse = process.stderr duration_match = re.search(r'Duration: (\\d{2}):(\\d{2}):(\\d{2}\\.\\d+)', output_to_parse) if duration_match: h, m, s = map(float, duration_match.groups()) duration = h * 3600 + m * 60 + s logging.info(f"[AlignContent] FFmpeg CLI成功获取duration: {duration:.2f}秒") else: # Try stdout duration_match_stdout = re.search(r'Duration: (\\d{2}):(\\d{2}):(\\d{2}\\.\\d+)', process.stdout) if duration_match_stdout: h, m, s = map(float, duration_match_stdout.groups()) duration = h * 3600 + m * 60 + s logging.info(f"[AlignContent] FFmpeg CLI (stdout)成功获取duration: {duration:.2f}秒") else: logging.error("[AlignContent] FFmpeg CLI无法获取duration。将使用timestamps。") except Exception as ff_exc_align: logging.error(f"[AlignContent] FFmpeg CLI获取duration时出错: {ff_exc_align}。将使用timestamps。") if duration <= 0: # If all attempts to get duration failed if timestamps and len(timestamps) > 0 : duration = timestamps[-1] + FRAME_INTERVAL # Estimate from last keyframe timestamp logging.warning(f"[AlignContent] 无法获取精确视频时长，根据最后关键帧估算为: {duration:.2f}秒") else: # Cannot determine duration at all logging.error("[AlignContent] 无法确定视频时长，内容对齐可能不准确。") # Potentially raise error or return empty if duration is critical # For now, proceed with duration=0, which might lead to issues in generate_page_intervals pass # Let it proceed with duration = 0, subsequent logic must handle # 语音识别 segments = VideoProcessor.transcribe_audio(video_path) if not segments: logging.warning("未识别到语音内容，将生成空文本摘要") segments = [] else: original_segment_count = len(segments) segments = ContentAligner._filter_repetitive_segments(segments) # 调用文本去重 if len(segments) < original_segment_count: logging.info(f"语音片段去重处理完成：从 {original_segment_count} 个片段减少到 {len(segments)} 个片段。") else: logging.info("语音片段无需去重处理。") # 生成页面时间间隔 intervals = ContentAligner.generate_page_intervals(timestamps, duration) # 使用改进的匹配算法（并行处理） page_texts = ContentAligner.find_best_match(segments, intervals) # 生成最终的对齐数据 aligned_data = [] for idx in range(len(intervals)): text = " ".join([seg["text"] for seg in page_texts.get(idx, [])]) aligned_data.append({ "page": idx, "start_time": intervals[idx][0], "end_time": intervals[idx][1], "text": text if text else "未识别到相关语音内容" }) # 停止超时处理 timeout_handler_align.stop() return aligned_data except Exception as e: logging.error(f"内容对齐失败: {str(e)}") logging.error(traceback.format_exc()) return [] finally: if timeout_handler_align: timeout_handler_align.stop() if cap_align and cap_align.isOpened(): cap_align.release() logging.info("[AlignContent] OpenCV VideoCapture released.") # ---------------------- 摘要生成模块 ---------------------- class SummaryGenerator: @staticmethod def optimize_text(text: str) -> str: """优化文本内容""" # 替换专业术语 for term, abbr in PROFESSIONAL_TERMS.items(): text = text.replace(term, f'{term} ({abbr})') # 优化过渡词 for word in TRANSITION_WORDS: text = text.replace(word, f'{word}') return text @staticmethod def generate_html(aligned_data: list, keyframes: list, output_dir: str): """生成HTML格式的报告""" # 创建临时目录用于存储图片 temp_img_dir = os.path.join(output_dir, "temp_images") os.makedirs(temp_img_dir, exist_ok=True) # 创建进度跟踪器 progress = ProgressTracker(len(aligned_data) + 1, "HTML报告生成") # 创建超时处理器 timeout_handler = TimeoutHandler() timeout_handler.start("HTML报告生成") try: # 检查输出目录权限 try: # 尝试在输出目录创建测试文件以验证权限 test_file = os.path.join(output_dir, "test_write_permission.tmp") with open(test_file, 'w') as f: f.write("test") os.remove(test_file) logging.info(f"输出目录权限检查通过: {output_dir}") except Exception as e: logging.error(f"输出目录权限检查失败: {str(e)}，尝试使用当前目录") # 如果指定的输出目录不可写，则使用当前目录 output_dir = os.path.abspath(".") temp_img_dir = os.path.join(output_dir, "temp_images") os.makedirs(temp_img_dir, exist_ok=True) logging.info(f"已切换到当前目录作为输出: {output_dir}") # 性能优化：减小图片大小，加快处理 logging.info("优化图片尺寸以提高性能") optimized_keyframes = [] for frame in keyframes: # 限制图片最大尺寸为720p if frame.width > 1280 or frame.height > 720: aspect_ratio = frame.width / frame.height if aspect_ratio > 16/9: # 宽屏 new_width = 1280 new_height = int(new_width / aspect_ratio) else: new_height = 720 new_width = int(new_height * aspect_ratio) frame = frame.resize((new_width, new_height), Image.LANCZOS) optimized_keyframes.append(frame) keyframes = optimized_keyframes logging.info("图片尺寸优化完成") # 处理所有帧 pages_data = [] for idx, frame in enumerate(keyframes): try: page_data = SummaryGenerator._process_frame(idx, frame, aligned_data, temp_img_dir) if page_data: pages_data.append(page_data) progress.update(1, f"处理第 {idx + 1} 页") except Exception as e: logging.error(f"处理帧 {idx} 时出错: {str(e)}") logging.error(traceback.format_exc()) continue # 检查是否有成功处理的页面 if not pages_data: logging.error("没有成功处理任何页面，无法生成HTML报告") raise RuntimeError("没有成功处理任何页面，无法生成HTML报告") # 生成HTML模板 template = Environment().from_string(""" PPT视频结构化摘要

PPT视频结构化摘要

{% for page in pages %}
第 {{ page.num }} 页
时间区间：{{ page.time }}
页面截图
{{ page.text }}
{% endfor %} """) # 保存HTML文件 output_path = os.path.join(output_dir, "summary.html") try: with open(output_path, "w", encoding="utf-8") as f: f.write(template.render(pages=pages_data)) logging.info(f"HTML报告已生成: {output_path}") # 检查文件是否已成功写入 if os.path.exists(output_path) and os.path.getsize(output_path) > 0: logging.info(f"HTML报告验证成功: {output_path}，大小: {os.path.getsize(output_path)} 字节") else: logging.error(f"HTML报告生成失败: 文件不存在或为空: {output_path}") raise IOError(f"HTML报告生成失败: 文件不存在或为空: {output_path}") except Exception as e: logging.error(f"HTML报告保存失败: {str(e)}") # 尝试使用备用路径 backup_path = os.path.join(os.path.abspath("."), f"summary_{int(time.time())}.html") logging.info(f"尝试使用备用路径保存HTML: {backup_path}") with open(backup_path, "w", encoding="utf-8") as f: f.write(template.render(pages=pages_data)) logging.info(f"HTML报告已使用备用路径生成: {backup_path}") output_path = backup_path # 更新输出路径 # 停止超时处理 timeout_handler.stop() progress.complete(f"HTML报告生成完成: {output_path}") # 打印明确的文件位置信息以便用户查找 print(f"\n[重要] HTML报告已生成在: {os.path.abspath(output_path)}\n") except Exception as e: logging.error(f"HTML报告生成过程中发生错误: {str(e)}") try: logging.error(traceback.format_exc()) except Exception: logging.error("无法获取详细错误信息，traceback模块不可用") # 停止超时处理 timeout_handler.stop() raise finally: # 清理临时文件 try: if os.path.exists(temp_img_dir): for f in os.listdir(temp_img_dir): try: os.remove(os.path.join(temp_img_dir, f)) except Exception as e: logging.error(f"删除临时图片文件失败: {str(e)}") try: os.rmdir(temp_img_dir) logging.info("已删除临时图片目录") except Exception as e: logging.error(f"删除临时图片目录失败: {str(e)}") except Exception as e: logging.error(f"清理临时文件时出错: {str(e)}") return output_path # 返回生成的HTML文件路径 @staticmethod def _process_frame(idx, frame, aligned_data, temp_img_dir): """处理单个帧""" try: img_path = os.path.join(temp_img_dir, f"page_{idx}.jpg") frame.save(img_path, quality=85) with open(img_path, "rb") as f: img_data = base64.b64encode(f.read()).decode("utf-8") return { "num": idx + 1, "time": f"{aligned_data[idx]['start_time']:.1f}s - {aligned_data[idx]['end_time']:.1f}s", "image": f"data:image/jpeg;base64,{img_data}", "text": SummaryGenerator.optimize_text(aligned_data[idx]["text"]) } except Exception as e: logging.error(f"处理帧 {idx} 时出错: {str(e)}") return None @staticmethod def generate_pdf(aligned_data: list, keyframes: list, output_dir: str): """生成PDF格式的报告""" # 首先生成HTML文件 html_path = os.path.join(output_dir, "summary.html") if not os.path.exists(html_path): SummaryGenerator.generate_html(aligned_data, keyframes, output_dir) # 创建进度跟踪器 progress = ProgressTracker(1, "PDF报告生成") # 创建超时处理器 timeout_handler = TimeoutHandler() timeout_handler.start("PDF报告生成") try: logging.info("开始将HTML转换为PDF...") # 设置PDF配置选项 options = { 'page-size': 'A4', 'margin-top': '0.75in', 'margin-right': '0.75in', 'margin-bottom': '0.75in', 'margin-left': '0.75in', 'encoding': 'UTF-8', 'no-outline': None, 'quiet': '' } # 生成PDF文件路径 pdf_path = os.path.join(output_dir, "summary.pdf") # 使用pdfkit生成PDF try: pdfkit.from_file(html_path, pdf_path, options=options) logging.info(f"PDF报告已生成: {pdf_path}") # 停止超时处理 timeout_handler.stop() progress.complete("PDF报告生成完成") return True except Exception as e: logging.error(f"PDF生成失败: {str(e)}") return False except Exception as e: logging.error(f"PDF报告生成过程出错: {str(e)}") timeout_handler.stop() return False @classmethod def generate_all(cls, aligned_data: list, keyframes: list, output_dir: str): """生成所有格式报告""" try: # 首先生成HTML报告 html_path = cls.generate_html(aligned_data, keyframes, output_dir) # 输出明确的报告位置提示 print(f"\n[完成] 报告生成成功！\n") print(f"HTML报告地址: {os.path.abspath(html_path)}") # 尝试生成PDF报告 pdf_success = False try: # 检查pdfkit模块是否可用 if 'pdfkit' in sys.modules: pdf_success = cls.generate_pdf(aligned_data, keyframes, output_dir) else: logging.info("pdfkit模块不可用，跳过PDF生成") except Exception as e: logging.error(f"PDF报告生成失败: {str(e)}") if not pdf_success: logging.warning("PDF生成功能不可用或生成失败，仅生成HTML报告") return True except Exception as e: logging.error(f"报告生成出错: {str(e)}") logging.error(traceback.format_exc()) # 创建一个极简的报告，以确保用户至少能看到一些结果 try: fallback_path = os.path.join(os.path.abspath("."), "emergency_report.html") with open(fallback_path, "w", encoding="utf-8") as f: f.write(f""" 应急报告

视频处理完成，但报告生成失败

处理过程中发生了以下错误:

{str(e)}

请查看日志文件以获取更多信息。

""") print(f"\n[警告] 正常报告生成失败，已创建应急报告: {fallback_path}\n") except Exception: logging.error("创建应急报告也失败了") return False # ---------------------- 主流程控制 ---------------------- def main_process(video_path, output_dir=None, progress_callback=None): try: logging.info(f"开始处理视频文件: {video_path}") # 设置输出目录 if output_dir is None: output_dir = OUTPUT_DIR # 检查输出目录是否存在，如果不存在则创建 try: os.makedirs(output_dir, exist_ok=True) logging.info(f"使用输出目录: {output_dir}") # 检查输出目录权限 test_file = os.path.join(output_dir, "test_permission.tmp") with open(test_file, "w") as f: f.write("test") os.remove(test_file) except Exception as e: logging.error(f"输出目录异常: {str(e)}，使用当前目录作为替代") output_dir = os.path.abspath(".") os.makedirs(output_dir, exist_ok=True) logging.info(f"已切换到当前目录: {output_dir}") # 进度回调函数 def update_progress(progress, message=None): if progress_callback: try: progress_callback(progress, message) except Exception as e: logging.error(f"进度回调函数执行失败: {str(e)}") logging.info(f"处理进度: {progress}% - {message if message else ''}") # 初始化进度 update_progress(0, "开始处理视频") # 检查视频文件是否存在 if not os.path.exists(video_path): error_msg = f"视频文件不存在: {video_path}" logging.error(error_msg) update_progress(0, f"错误: {error_msg}") raise FileNotFoundError(error_msg) # 检查文件大小 file_size = os.path.getsize(video_path) / (1024 * 1024) # 转换为MB logging.info(f"视频文件大小: {file_size:.2f}MB") # 检查文件是否为空 if file_size == 0: error_msg = "视频文件为空" logging.error(error_msg) update_progress(0, f"错误: {error_msg}") raise ValueError(error_msg) # 检查文件是否可读 try: with open(video_path, 'rb') as f: f.read(1024) # 尝试读取一小块数据 except Exception as e: error_msg = f"视频文件无法读取: {str(e)}" logging.error(error_msg) update_progress(0, f"错误: {error_msg}") raise IOError(error_msg) # 检查依赖项 update_progress(5, "检查系统依赖") if not check_dependencies(): error_msg = "依赖项检查失败" logging.error(error_msg) update_progress(5, f"错误: {error_msg}") raise RuntimeError(f"{error_msg}，请检查日志获取详细信息") update_progress(10, "依赖项检查通过") # 初始化视频处理器 processor = VideoProcessor() # 提取关键帧 logging.info("开始提取关键帧...") update_progress(15, "开始提取关键帧") try: keyframes, duration = processor.extract_keyframes(video_path) if not keyframes: error_msg = "关键帧提取失败：未能提取到任何关键帧" logging.error(error_msg) update_progress(15, f"错误: 未能提取到关键帧") raise RuntimeError(error_msg) logging.info(f"成功提取 {len(keyframes)} 个关键帧，视频时长：{duration:.2f}秒") update_progress(40, f"已提取 {len(keyframes)} 个关键帧") except Exception as e: error_msg = f"关键帧提取过程出错: {str(e)}" logging.error(error_msg) try: logging.error(traceback.format_exc()) except Exception: logging.error("无法获取详细错误信息，traceback模块不可用") update_progress(15, f"错误: 关键帧提取失败 - {str(e)}") raise RuntimeError(error_msg) # 转录音频 logging.info("开始转录音频...") update_progress(45, "开始转录音频") try: segments = VideoProcessor.transcribe_audio(video_path) if not segments: logging.warning("音频转录失败：未能识别到任何语音内容") update_progress(45, "警告: 未识别到语音内容，将生成空文本摘要") segments = [] else: logging.info(f"成功转录 {len(segments)} 个音频片段") update_progress(65, f"已转录 {len(segments)} 个音频片段") for i, seg in enumerate(segments[:3], 1): # 只记录前三个片段作为示例 logging.debug(f"音频片段 {i}: {seg['text'][:50]}...") except Exception as e: error_msg = f"音频转录过程出错: {str(e)}" logging.error(error_msg) try: logging.error(traceback.format_exc()) except Exception: logging.error("无法获取详细错误信息，traceback模块不可用") update_progress(45, f"错误: 音频转录失败 - {str(e)}") raise RuntimeError(error_msg) # 计算时间戳 timestamps = [0] # 添加起始时间戳 for frame_idx, frame in enumerate(keyframes[1:], 1): timestamps.append(frame_idx * duration / len(keyframes)) # 对齐内容 logging.info("开始对齐内容...") update_progress(70, "开始对齐内容") try: aligned_data = ContentAligner.align_content(video_path, timestamps) if not aligned_data: error_msg = "内容对齐失败：未能生成对齐数据" logging.error(error_msg) update_progress(70, "错误: 内容对齐失败") # 创建一个空的对齐数据，以便能继续生成报告 aligned_data = [] for i in range(len(keyframes)): aligned_data.append({ "page": i, "start_time": timestamps[i], "end_time": timestamps[i+1] if i < len(timestamps)-1 else duration, "text": "未能识别到相关语音内容" }) logging.info(f"已创建{len(aligned_data)}个空内容对齐数据") update_progress(75, "使用空内容继续处理") else: logging.info(f"成功对齐 {len(aligned_data)} 个内容片段") update_progress(80, f"已对齐 {len(aligned_data)} 个内容片段") for i, data in enumerate(aligned_data[:3], 1): # 只记录前三个对齐结果作为示例 logging.debug(f"对齐片段 {i}: {data.get('start_time', 'N/A')}s - {data.get('end_time', 'N/A')}s") except Exception as e: error_msg = f"内容对齐过程出错: {str(e)}" logging.error(error_msg) try: logging.error(traceback.format_exc()) except Exception: logging.error("无法获取详细错误信息，traceback模块不可用") update_progress(70, f"错误: 内容对齐失败 - {str(e)}") # 创建一个空的对齐数据，以便能继续生成报告 aligned_data = [] for i in range(len(keyframes)): aligned_data.append({ "page": i, "start_time": timestamps[i], "end_time": timestamps[i+1] if i < len(timestamps)-1 else duration, "text": "未能识别到相关语音内容" }) logging.info(f"已创建{len(aligned_data)}个空内容对齐数据") update_progress(75, "使用空内容继续处理") # 生成总结 logging.info("开始生成总结...") update_progress(85, "开始生成报告") try: if SummaryGenerator.generate_all(aligned_data, keyframes, output_dir): logging.info(f"总结生成完成，输出目录: {output_dir}") update_progress(100, "处理完成") # 检查HTML文件是否存在 html_path = os.path.join(output_dir, "summary.html") if os.path.exists(html_path): logging.info(f"报告验证成功: {html_path}") print(f"\n[成功] 报告生成完成，位置: {os.path.abspath(html_path)}\n") else: logging.warning(f"报告文件不存在: {html_path}") print(f"\n[警告] 处理似乎完成但未找到报告文件，请检查日志\n") else: error_msg = "报告生成失败" logging.error(error_msg) update_progress(85, f"错误: {error_msg}") raise RuntimeError(error_msg) except Exception as e: error_msg = f"总结生成过程出错: {str(e)}" logging.error(error_msg) try: logging.error(traceback.format_exc()) except Exception: logging.error("无法获取详细错误信息，traceback模块不可用") update_progress(85, f"错误: 报告生成失败 - {str(e)}") # 尝试创建一个简单的报告 try: simple_html = os.path.join(output_dir, "simple_report.html") with open(simple_html, "w", encoding="utf-8") as f: f.write(f""" 简单报告

视频简单报告

完整报告生成失败，这是一个简化版本

""") # 添加关键帧 for i, frame in enumerate(keyframes): # 保存图片 img_path = os.path.join(output_dir, f"frame_{i}.jpg") frame.save(img_path) # 添加到HTML f.write(f"""

第 {i+1} 帧

关键帧 {i+1}
""") f.write("") logging.info(f"简单报告已生成: {simple_html}") print(f"\n[恢复] 创建了简单报告: {os.path.abspath(simple_html)}\n") except Exception as inner_e: logging.error(f"简单报告生成也失败了: {str(inner_e)}") raise RuntimeError(error_msg) logging.info("所有处理步骤已完成") return True except Exception as e: logging.error(f"处理过程中发生错误: {str(e)}") logging.error("详细错误信息:") try: logging.error(traceback.format_exc()) except Exception: logging.error("无法获取详细错误信息，traceback模块不可用") if progress_callback: try: progress_callback(0, f"处理失败: {str(e)}") except: pass print(f"\n[错误] 处理失败: {str(e)}\n") return False if __name__ == "__main__": try: if len(sys.argv) < 2: print("使用方法: python 毕设.py <视频文件路径>") sys.exit(1) video_path = sys.argv[1] if main_process(video_path): print("[完成] 处理成功") sys.exit(0) else: print("[错误] 处理失败，请查看日志文件了解详情") sys.exit(1) except KeyboardInterrupt: print("\n[中断] 用户中断了处理") sys.exit(130) except Exception as e: print(f"[错误] 程序执行过程中出现未处理的异常: {str(e)}") try: traceback.print_exc() except Exception: print("无法打印详细错误信息，traceback模块不可用") sys.exit(1)