From aa96ed1bbec4d578d4c737e468b762b1e83d6c48 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E9=BB=84=E5=87=AF?= <2684756428@qq.com> Date: Sun, 21 Dec 2025 21:44:09 +0800 Subject: [PATCH 1/4] Enhance config.py with environment variable support Refactor configuration management to support environment variables and improve readability. --- .../config.py | 99 +++++++++++++++---- 1 file changed, 79 insertions(+), 20 deletions(-) diff --git a/src/Driving_Accident_Video_Recognition/config.py b/src/Driving_Accident_Video_Recognition/config.py index 78f37f238b..d93413cbc7 100644 --- a/src/Driving_Accident_Video_Recognition/config.py +++ b/src/Driving_Accident_Video_Recognition/config.py @@ -1,29 +1,88 @@ """ -全局配置文件:所有可配置参数集中管理,新手只需修改这里 +配置模块:支持环境变量/`.env`加载 + 精细化事故判断配置 +优化点:健壮性提升+代码简洁性+可读性增强 """ -# YOLOv8模型配置 -YOLO_MODEL_PATH = "yolov8n.pt" # 轻量化模型(自动下载) -CONFIDENCE_THRESHOLD = 0.5 # 目标检测置信度阈值(0-1) +import os +from typing import Any +from dotenv import load_dotenv -# 检测源配置 -DETECTION_SOURCE = 0 # 0=电脑摄像头,可改为视频路径如"test_accident.mp4" +# 加载.env环境文件(优先读取环境变量,无则用代码默认值) +load_dotenv() -# 事故识别配置 -ACCIDENT_CLASSES = [0, 2, 7] # YOLOv8类别:0=人,2=汽车,7=卡车 -MIN_VEHICLE_COUNT = 2 # 至少2辆车判定为事故 -PERSON_VEHICLE_CONTACT = True # 行人和车辆同时出现判定为事故 -# 帧处理优化配置(低配电脑推荐) -RESIZE_WIDTH = 640 -RESIZE_HEIGHT = 480 +def get_env_config(key: str, default: Any, config_type: type) -> Any: + """ + 统一处理环境变量读取+类型转换(减少重复代码,提升健壮性) + :param key: 环境变量键名 + :param default: 类型匹配的默认值 + :param config_type: 目标类型(int/float/bool/str等) + :return: 转换后的配置值(转换失败自动回退到默认值) + """ + env_value = os.getenv(key) + if env_value is None: + return default + try: + if config_type == bool: + # 布尔类型特殊处理:"True"/"true"转True,其余转False + return env_value.strip().lower() == "true" + return config_type(env_value) + except (ValueError, TypeError): + # 类型转换失败时,回退到默认值 + return default -# 依赖包配置 + +# ====================== YOLO模型配置 ====================== +# YOLO预训练模型路径(默认轻量型yolov8n.pt,适合实时检测) +YOLO_MODEL_PATH = get_env_config("YOLO_MODEL_PATH", "yolov8n.pt", str) +# 检测置信度阈值(0-1,值越高检测越严格,默认0.5平衡精度与召回) +CONFIDENCE_THRESHOLD = get_env_config("CONFIDENCE_THRESHOLD", 0.5, float) + + +# ====================== 检测源配置 ====================== +# 检测源:支持摄像头(整数设备号)或视频文件路径(字符串) +DETECTION_SOURCE = os.getenv("DETECTION_SOURCE", "0") +try: + # 尝试转换为整数(对应摄像头设备号,如0=默认摄像头) + DETECTION_SOURCE = int(DETECTION_SOURCE) +except ValueError: + # 转换失败则视为视频文件路径(保持字符串) + pass + + +# ====================== 事故识别核心配置 ====================== +# 事故检测关注的YOLO类别(0=person/行人、2=car/汽车、7=truck/卡车) +ACCIDENT_CLASSES = [0, 2, 7] +# 多车事故判定阈值:至少检测到N辆车辆才判定为多车事故 +MIN_VEHICLE_COUNT = get_env_config("MIN_VEHICLE_COUNT", 2, int) +# 是否开启“人车接触”事故判定(True=开启,检测到行人+车辆即判定) +PERSON_VEHICLE_CONTACT = get_env_config("PERSON_VEHICLE_CONTACT", True, bool) +# 人车接触距离阈值(像素):行人和车辆框中心距离<该值时,判定为接触 +PERSON_VEHICLE_DISTANCE_THRESHOLD = get_env_config( + "PERSON_VEHICLE_DISTANCE_THRESHOLD", 50, int +) + + +# ====================== 帧处理配置(平衡速度与精度) ====================== +# 检测帧缩放宽度(默认640,YOLO推荐输入尺寸,兼顾速度) +RESIZE_WIDTH = get_env_config("RESIZE_WIDTH", 640, int) +# 检测帧缩放高度(默认480,与宽度配合保持合理比例) +RESIZE_HEIGHT = get_env_config("RESIZE_HEIGHT", 480, int) + + +# ====================== 依赖包配置(自动安装时使用) ====================== REQUIRED_PACKAGES = [ - "ultralytics>=8.0.0", - "opencv-python>=4.8.0", - "numpy>=1.24.0", - "torch>=2.0.0" + "ultralytics>=8.0.0", # YOLOv8核心依赖 + "opencv-python>=4.8.0", # 视频/图像读取、绘制标注 + "numpy>=1.24.0", # 数值计算(坐标/距离运算) + "torch>=2.0.0", # YOLO模型推理(PyTorch后端) + "python-dotenv>=1.0.0" # 加载.env环境变量 ] +# PyPI镜像源(加速国内环境的依赖安装) +PYPI_MIRROR = "https://pypi.tuna.tsinghua.edu.cn/simple" + -# 清华镜像源(加速依赖下载) -PYPI_MIRROR = "https://pypi.tuna.tsinghua.edu.cn/simple" \ No newline at end of file +# ====================== 检测结果输出配置 ====================== +# 是否保存检测结果视频(True=保存,False=不保存) +SAVE_RESULT_VIDEO = get_env_config("SAVE_RESULT_VIDEO", False, bool) +# 检测结果视频保存路径(默认输出到项目根目录) +RESULT_VIDEO_PATH = get_env_config("RESULT_VIDEO_PATH", "detection_result.mp4", str) From 65d544208aeb5b0229336f6e3bd48fc1cf4d5101 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E9=BB=84=E5=87=AF?= <2684756428@qq.com> Date: Mon, 22 Dec 2025 10:12:17 +0800 Subject: [PATCH 2/4] Refactor accident video detection tool with enhancements Refactor main program for accident video detection with performance improvements, flexible configurations, and enhanced logging. Added support for recognizing people and vehicles. --- .../main.py | 114 ++++++++++++++---- 1 file changed, 92 insertions(+), 22 deletions(-) diff --git a/src/Driving_Accident_Video_Recognition/main.py b/src/Driving_Accident_Video_Recognition/main.py index 073a804427..8419e7e63b 100644 --- a/src/Driving_Accident_Video_Recognition/main.py +++ b/src/Driving_Accident_Video_Recognition/main.py @@ -1,41 +1,111 @@ """ -主程序:驾驶事故视频识别工具 +主程序:驾驶事故视频识别工具(优化版) +优化点:性能提速+灵活配置+规范日志+新增人和小车识别提示 """ import sys import os import argparse +import logging # 新增:日志模块(替代print,支持分级输出) +from config import ( + REQUIRED_PACKAGES, PYPI_MIRROR, DETECTION_SOURCE, + CONFIDENCE_THRESHOLD, ACCIDENT_CLASSES # 新增:引入识别类别配置 +) +from utils.dependencies import install_dependencies +from core.detector import AccidentDetector -# 确保当前目录可被搜索 -current_dir = os.path.dirname(os.path.abspath(__file__)) -sys.path.append(current_dir) +# -------------------------- 新增1:日志初始化(替代print,更灵活) -------------------------- +def init_logger(): + logger = logging.getLogger("AccidentDetection") + logger.setLevel(logging.INFO) + # 控制台输出格式:时间+日志级别+内容 + formatter = logging.Formatter("%(asctime)s - %(levelname)s - %(message)s") + console_handler = logging.StreamHandler() + console_handler.setFormatter(formatter) + logger.addHandler(console_handler) + return logger -# 直接导入同目录的文件(彻底避免模块包问题) -from config import REQUIRED_PACKAGES, PYPI_MIRROR, PERSON_VEHICLE_DISTANCE, ACCIDENT_CONTINUOUS_FRAMES -from dependencies import install_dependencies # 直接导入同目录的dependencies.py -from detector import AccidentDetector +# -------------------------- 新增2:优化命令行参数(更灵活的配置) -------------------------- +def parse_args(logger): + parser = argparse.ArgumentParser(description="驾驶事故视频识别工具(支持动态配置)") + # 基础参数:检测源、语言 + parser.add_argument("--source", "-s", default=DETECTION_SOURCE, + help=f"检测源(0=摄像头/视频路径,默认:{DETECTION_SOURCE})") + parser.add_argument("--language", "-l", default="zh", choices=["zh", "en"], + help="标注语言(zh=中文/en=英文,默认:zh)") + # 新增:性能/配置参数(无需改config.py,直接命令行调整) + parser.add_argument("--skip-deps", "-sd", action="store_true", default=False, + help="跳过依赖检查(已安装依赖时用,提速)") + parser.add_argument("--conf", "-c", type=float, default=CONFIDENCE_THRESHOLD, + help=f"检测置信度阈值(0-1,默认:{CONFIDENCE_THRESHOLD})") + # 新增:日志级别(调试/正常模式切换) + parser.add_argument("--log-level", "-ll", default="INFO", choices=["DEBUG", "INFO", "WARNING"], + help="日志级别(DEBUG=调试/INFO=正常/WARNING=仅警告,默认:INFO)") + + args = parser.parse_args() + # 校验参数合法性(新增:避免无效输入) + if not (0 < args.conf <= 1): + logger.warning(f"置信度{args.conf}无效,自动使用默认值{CONFIDENCE_THRESHOLD}") + args.conf = CONFIDENCE_THRESHOLD + return args +# -------------------------- 优化3:主函数逻辑(减少重复计算+提升健壮性+新增人和小车识别) -------------------------- +def main(): + # 初始化日志 + logger = init_logger() + # 解析参数(并应用日志级别) + args = parse_args(logger) + logger.setLevel(args.log_level) # 动态调整日志级别 -def parse_args(): - parser = argparse.ArgumentParser(description="驾驶事故视频识别") - parser.add_argument("--source", "-s", default=0, help="检测源:0=摄像头/视频路径") - parser.add_argument("--language", "-l", default="zh", choices=["zh", "en"], help="标注语言") - return parser.parse_args() + # -------------------------- 优化4:缓存环境变量操作(减少属性查找,提速) -------------------------- + env = os.environ # 局部变量缓存os.environ,避免循环中重复查找(参考摘要5“缓存属性”) + # 覆盖检测源(命令行优先) + if str(args.source) != str(DETECTION_SOURCE): + # 严谨处理检测源类型:尝试转整数(摄像头),失败则为字符串(视频路径) + try: + env["DETECTION_SOURCE"] = str(int(args.source)) # 摄像头(数字) + except (ValueError, TypeError): + env["DETECTION_SOURCE"] = str(args.source) # 视频路径(字符串) + logger.info(f"检测源已覆盖为:{env['DETECTION_SOURCE']}") + # 覆盖置信度阈值(命令行优先) + if args.conf != CONFIDENCE_THRESHOLD: + env["CONFIDENCE_THRESHOLD"] = str(args.conf) + logger.info(f"置信度阈值已覆盖为:{args.conf}") -def main(): - args = parse_args() try: - print("🚀 启动驾驶事故检测...") - # 安装依赖 - install_dependencies(REQUIRED_PACKAGES, PYPI_MIRROR) - # 启动检测 + logger.info("🚀 启动驾驶事故视频识别工具...") + # -------------------------- 优化5:跳过依赖检查(避免重复安装,提速) -------------------------- + if not args.skip_deps: + install_dependencies(REQUIRED_PACKAGES, PYPI_MIRROR) + else: + logger.info("⚠️ 已跳过依赖检查(--skip-deps生效)") + + # -------------------------- 优化6:简化检测器初始化(减少冗余代码) -------------------------- + logger.info("🔄 初始化事故检测器...") detector = AccidentDetector() + # 新增:提示当前模型支持识别人和小车 + target_classes = {0: "人", 2: "小车"} + supported_targets = [f"{name}(类别ID: {cid})" for cid, name in target_classes.items() if cid in ACCIDENT_CLASSES] + logger.info(f"✅ 检测器初始化完成,当前模型支持识别:{', '.join(supported_targets)}") + logger.info("✅ 开始检测(按Q/ESC退出,画面中会标注识别到的人和小车)") + + # 启动检测(传递语言参数) detector.run_detection(language=args.language) + except KeyboardInterrupt: - print("\n🛑 程序中断") + logger.info("\n🛑 用户强制中断程序") + except Exception as e: + # 新增:DEBUG级别输出详细异常栈,INFO级别只显示错误信息(方便调试) + logger.error(f"\n❌ 程序运行出错:{str(e)}") + if args.log_level == "DEBUG": + import traceback + traceback.print_exc() finally: - print("👋 程序退出") - + logger.info("👋 程序正常退出") if __name__ == "__main__": + # 新增:确保code目录在搜索路径(兼容不同运行方式) + current_dir = os.path.dirname(os.path.abspath(__file__)) + if current_dir not in sys.path: + sys.path.append(current_dir) main() From 6f88d57b0c4aa49e722d6bbbd818d56887cb8f91 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E9=BB=84=E5=87=AF?= <2684756428@qq.com> Date: Mon, 22 Dec 2025 10:33:47 +0800 Subject: [PATCH 3/4] Refactor process.py for improved functionality Refactor processing functions for accident video recognition, adding distance calculation and box center retrieval. --- .../process.py | 84 +++++++++++-------- 1 file changed, 48 insertions(+), 36 deletions(-) diff --git a/src/Driving_Accident_Video_Recognition/process.py b/src/Driving_Accident_Video_Recognition/process.py index 87df8b1bbd..cba31e2bc1 100644 --- a/src/Driving_Accident_Video_Recognition/process.py +++ b/src/Driving_Accident_Video_Recognition/process.py @@ -1,46 +1,58 @@ """ -辅助处理工具:负责坐标转换、帧处理等通用逻辑 +process.py:辅助函数(处理坐标、距离、标注) """ -import numpy as np -import torch import cv2 +import numpy as np + def process_box_coords(box, scale_x, scale_y): + """将YOLO输出的坐标缩放回原始帧尺寸""" + x1, y1, x2, y2 = box.xyxy[0].tolist() + return ( + int(x1 * scale_x), + int(y1 * scale_y), + int(x2 * scale_x), + int(y2 * scale_y) + ) + + +def get_box_center(x1, y1, x2, y2): + """计算目标框的中心坐标""" + return (int((x1 + x2) / 2), int((y1 + y2) / 2)) + + +def calculate_euclidean_distance(pt1, pt2): + """计算两个点的欧式距离""" + return np.linalg.norm(np.array(pt1) - np.array(pt2)) + + +def draw_annotations(frame, detected_objects, is_accident, language="zh"): """ - 安全处理YOLOv8检测框坐标,解决张量类型错误 - :param box: YOLOv8的检测框对象 - :param scale_x: 宽度缩放比例 - :param scale_y: 高度缩放比例 - :return: 转换后的坐标(x1, y1, x2, y2) - """ - # 兼容张量和numpy数组 - if isinstance(box.xyxy[0], torch.Tensor): - box_xyxy = box.xyxy[0].cpu().numpy() - else: - box_xyxy = np.array(box.xyxy[0]) - # 缩放坐标 - scaled_box = box_xyxy * [scale_x, scale_y, scale_x, scale_y] - # 转换为整数 - return map(int, scaled_box) - -def draw_annotations(frame, detected_objects, is_accident): - """ - 在帧上绘制检测标注和事故警告 - :param frame: 原始视频帧 - :param detected_objects: 检测到的目标列表 - :param is_accident: 是否检测到事故 - :return: 标注后的帧 + 绘制标注(无需扩展库,用拼音+中文注释避免乱码) """ - # 绘制目标检测框 - for (cls_name, x1, y1, x2, y2) in detected_objects: + # 类别映射:拼音+中文(OpenCV默认支持英文/拼音) + class_map = { + "person": "Ren(人)" if language == "zh" else "Person", + "car": "Xiao Che(小车)" if language == "zh" else "Car", + "truck": "Ka Che(卡车)" if language == "zh" else "Truck" + } + + # OpenCV默认英文字体(无需扩展库) + font = cv2.FONT_HERSHEY_SIMPLEX + + # 绘制目标框+标签 + for obj in detected_objects: + cls_name, x1, y1, x2, y2 = obj + display_name = class_map.get(cls_name, cls_name) + # 绘制绿色框 cv2.rectangle(frame, (x1, y1), (x2, y2), (0, 255, 0), 2) - cv2.putText(frame, cls_name, (x1, y1 - 10), - cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 0), 2) - # 绘制事故警告 + # 绘制标签(避免超出画面) + label_y = y1 - 10 if y1 > 20 else y1 + 20 + cv2.putText(frame, display_name, (x1, label_y), font, 0.8, (0, 255, 0), 2) + + # 绘制事故提示(红色) if is_accident: - cv2.putText(frame, "⚠️ 检测到事故!", (50, 50), - cv2.FONT_HERSHEY_SIMPLEX, 1.2, (0, 0, 255), 3) - return frame + accident_text = "Shi Gu!(事故!)" if language == "zh" else "Accident Detected!" + cv2.putText(frame, accident_text, (50, 50), font, 1.2, (0, 0, 255), 3) -# 供外部导入的函数 -__all__ = ["process_box_coords", "draw_annotations"] \ No newline at end of file + return frame From 8ec4153259005c12605508fe3b6a9caa8974d4b0 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E9=BB=84=E5=87=AF?= <2684756428@qq.com> Date: Mon, 22 Dec 2025 10:34:59 +0800 Subject: [PATCH 4/4] Refactor AccidentDetector for improved functionality Enhance accident detection with video saving and FPS display. --- .../detector.py | 193 ++++++++++++------ 1 file changed, 132 insertions(+), 61 deletions(-) diff --git a/src/Driving_Accident_Video_Recognition/detector.py b/src/Driving_Accident_Video_Recognition/detector.py index 4155840ab8..d281ccaf87 100644 --- a/src/Driving_Accident_Video_Recognition/detector.py +++ b/src/Driving_Accident_Video_Recognition/detector.py @@ -1,121 +1,192 @@ """ -事故检测器核心类:负责模型加载、检测流程执行 +检测器模块:精准事故判断+视频保存+帧率显示(无报错版) """ import sys import cv2 +import time from ultralytics import YOLO from config import ( YOLO_MODEL_PATH, CONFIDENCE_THRESHOLD, ACCIDENT_CLASSES, - MIN_VEHICLE_COUNT, PERSON_VEHICLE_CONTACT, - RESIZE_WIDTH, RESIZE_HEIGHT, DETECTION_SOURCE + MIN_VEHICLE_COUNT, PERSON_VEHICLE_CONTACT, PERSON_VEHICLE_DISTANCE_THRESHOLD, + RESIZE_WIDTH, RESIZE_HEIGHT, DETECTION_SOURCE, + SAVE_RESULT_VIDEO, RESULT_VIDEO_PATH ) -from core.process import process_box_coords, draw_annotations +from core.process import ( + process_box_coords, get_box_center, calculate_euclidean_distance, draw_annotations +) + class AccidentDetector: def __init__(self): - """初始化检测器,加载YOLOv8模型""" - self.model = None - self.accident_detected = False - self._load_model() + self.model = None # YOLO模型对象 + self.accident_detected = False # 是否检测到事故 + self.video_writer = None # 视频写入器(保存检测结果) + # 帧率计算(滑动平均,避免波动) + self.fps_history = [] + self.prev_time = time.time() + + self._load_model() # 初始化时加载模型 def _load_model(self): - """私有方法:加载模型,包含重试逻辑""" + """加载YOLO模型(增加兜底逻辑)""" + print("🔄 加载YOLOv8检测模型...") try: - print("🔄 正在加载YOLOv8模型(首次运行会自动下载)...") self.model = YOLO(YOLO_MODEL_PATH) - print("✅ YOLOv8模型加载成功") + print(f"✅ 模型加载成功:{YOLO_MODEL_PATH}") except Exception as e: - print(f"❌ 模型加载失败:{e}") - # 重试加载模型 + print(f"⚠️ 指定模型加载失败,尝试默认轻量模型yolov8n.pt...") try: - print("🔄 尝试重新下载模型...") self.model = YOLO("yolov8n.pt") - print("✅ 模型重新加载成功") + print("✅ 兜底模型(yolov8n.pt)加载成功") except Exception as e2: - print(f"❌ 模型重新加载失败:{e2}") + print(f"❌ 模型加载失败:{e2},程序退出") sys.exit(1) - def detect_frame(self, frame): - """处理单帧,返回标注后的帧和是否检测到事故""" + def _init_video_writer(self, frame): + """初始化视频写入器(增加路径检查)""" + if not SAVE_RESULT_VIDEO: + return + height, width = frame.shape[:2] + fourcc = cv2.VideoWriter_fourcc(*"mp4v") + # 自动创建保存目录(避免路径不存在) + save_dir = "/".join(RESULT_VIDEO_PATH.split("/")[:-1]) + if save_dir and not cv2.os.path.exists(save_dir): + cv2.os.makedirs(save_dir) + # 初始化写入器 + self.video_writer = cv2.VideoWriter(RESULT_VIDEO_PATH, fourcc, 30.0, (width, height)) + if not self.video_writer.isOpened(): + print(f"⚠️ 无法保存视频到{RESULT_VIDEO_PATH},跳过保存") + self.video_writer = None + + def _calculate_accident(self, detected_objects): + """精准判断事故:多车/人车接触""" + persons = [obj for obj in detected_objects if obj[0] == "person"] + vehicles = [obj for obj in detected_objects if obj[0] in ["car", "truck"]] + + # 条件1:车辆数量≥配置阈值 + if len(vehicles) >= MIN_VEHICLE_COUNT: + return True + # 条件2:行人和车辆距离≤阈值 + if PERSON_VEHICLE_CONTACT and len(persons) >= 1 and len(vehicles) >= 1: + p_centers = [get_box_center(*obj[1:]) for obj in persons] + v_centers = [get_box_center(*obj[1:]) for obj in vehicles] + for p in p_centers: + for v in v_centers: + if calculate_euclidean_distance(p, v) <= PERSON_VEHICLE_DISTANCE_THRESHOLD: + return True + return False + + def detect_frame(self, frame, language="zh"): + """处理单帧:检测+标注+帧率计算""" detected_objects = [] + current_frame = frame.copy() + try: - # 缩放帧提升速度 - frame_resized = cv2.resize(frame, (RESIZE_WIDTH, RESIZE_HEIGHT)) - # YOLOv8推理 - results = self.model(frame_resized, conf=CONFIDENCE_THRESHOLD) + # 缩放帧(适配YOLO输入) + frame_resized = cv2.resize(current_frame, (RESIZE_WIDTH, RESIZE_HEIGHT)) + # 模型推理(关闭冗余日志) + results = self.model(frame_resized, conf=CONFIDENCE_THRESHOLD, verbose=False) # 解析检测结果 for r in results: - if hasattr(r, 'boxes') and r.boxes is not None: - for box in r.boxes: - if not hasattr(box, 'cls') or box.cls is None: - continue - cls_idx = int(box.cls[0]) - if cls_idx in ACCIDENT_CLASSES: - cls_name = self.model.names[cls_idx] - # 处理坐标 - scale_x = frame.shape[1] / RESIZE_WIDTH - scale_y = frame.shape[0] / RESIZE_HEIGHT - x1, y1, x2, y2 = process_box_coords(box, scale_x, scale_y) - detected_objects.append((cls_name, x1, y1, x2, y2)) + if not hasattr(r, "boxes") or r.boxes is None: + continue + for box in r.boxes: + if not hasattr(box, "cls") or box.cls is None: + continue + cls_idx = int(box.cls[0]) + if cls_idx in ACCIDENT_CLASSES: + cls_name = self.model.names[cls_idx] + # 坐标缩放回原始帧 + scale_x = current_frame.shape[1] / RESIZE_WIDTH + scale_y = current_frame.shape[0] / RESIZE_HEIGHT + x1, y1, x2, y2 = process_box_coords(box, scale_x, scale_y) + detected_objects.append((cls_name, x1, y1, x2, y2)) # 判断事故 - person_count = sum(1 for obj in detected_objects if obj[0] == "person") - vehicle_count = sum(1 for obj in detected_objects if obj[0] in ["car", "truck"]) - is_accident = (vehicle_count >= MIN_VEHICLE_COUNT) or (person_count >= 1 and vehicle_count >= 1 and PERSON_VEHICLE_CONTACT) - self.accident_detected = is_accident - + self.accident_detected = self._calculate_accident(detected_objects) # 绘制标注 - frame = draw_annotations(frame, detected_objects, is_accident) + current_frame = draw_annotations(current_frame, detected_objects, self.accident_detected, language) + + # 计算滑动平均帧率 + current_time = time.time() + self.fps_history.append(1 / (current_time - self.prev_time)) + self.prev_time = current_time + # 只保留最近10帧的帧率(避免波动) + if len(self.fps_history) > 10: + self.fps_history.pop(0) + avg_fps = int(sum(self.fps_history) / len(self.fps_history)) if self.fps_history else 0 + # 绘制帧率 + cv2.putText(current_frame, f"FPS: {avg_fps}", (50, 100), + cv2.FONT_HERSHEY_SIMPLEX, 1, (255, 0, 0), 2) + + # 保存视频帧 + if self.video_writer: + self.video_writer.write(current_frame) except Exception as e: - print(f"⚠️ 帧处理出现小错误:{e},继续运行...") + print(f"⚠️ 帧处理错误:{e},继续运行...") - return frame, self.accident_detected + return current_frame, self.accident_detected - def run_detection(self): - """启动检测流程,包含完善的容错逻辑""" - # 多次尝试打开检测源 + def run_detection(self, language="zh"): + """启动检测流程:打开摄像头/视频+逐帧处理""" + # 打开检测源(重试3次) cap = None - for i in range(3): + for retry in range(3): cap = cv2.VideoCapture(DETECTION_SOURCE) if cap.isOpened(): + print(f"✅ 第{retry+1}次打开检测源成功") break - print(f"⚠️ 第{i+1}次打开检测源失败,重试中...") - cv2.waitKey(1000) + print(f"⚠️ 第{retry+1}次打开检测源失败,1秒后重试...") + time.sleep(1) + # 兜底:打开默认摄像头 if not cap or not cap.isOpened(): - print(f"❌ 无法打开检测源:{DETECTION_SOURCE}") - # 强制切换为摄像头 - print("🔄 强制切换为电脑摄像头...") + print(f"❌ 目标检测源{DETECTION_SOURCE}无法打开,尝试默认摄像头(0)...") cap = cv2.VideoCapture(0) if not cap.isOpened(): - print("❌ 摄像头也无法打开,请检查设备") + print("❌ 所有检测源均无法打开,程序退出") sys.exit(1) - print("✅ 检测源打开成功,开始实时检测(按Q/ESC键退出)") - print("💡 提示:检测到2辆车或行人和车辆同时出现时,显示红色警告") + print("✅ 检测源打开成功(按Q/ESC退出)") + print(f"💡 配置:行人车辆距离阈值{PERSON_VEHICLE_DISTANCE_THRESHOLD}像素") + + # 初始化视频写入器(读取第一帧) + ret, first_frame = cap.read() + if ret: + self._init_video_writer(first_frame) # 逐帧处理 while True: ret, frame = cap.read() if not ret: - print("🔚 视频/摄像头流结束") + print("🔚 视频流读取完毕,结束检测") break - frame, _ = self.detect_frame(frame) - cv2.imshow("驾驶事故检测(按Q退出)", frame) + # 处理单帧 + processed_frame, _ = self.detect_frame(frame, language) + cv2.imshow("驾驶事故检测", processed_frame) # 退出逻辑 key = cv2.waitKey(1) & 0xFF - if key == ord('q') or key == 27: + if key == ord("q") or key == 27: print("🛑 用户手动退出") break # 释放资源 cap.release() + if self.video_writer: + self.video_writer.release() + print(f"✅ 检测结果已保存到{RESULT_VIDEO_PATH}") cv2.destroyAllWindows() - print(f"\n📊 检测总结:是否检测到事故 → {'✅ 是' if self.accident_detected else '❌ 否'}") -# 供外部导入的类 -__all__ = ["AccidentDetector"] \ No newline at end of file + # 检测总结 + avg_fps = int(sum(self.fps_history) / len(self.fps_history)) if self.fps_history else 0 + print(f"\n📊 检测总结:") + print(f" - 是否检测到事故 → {'✅ 是' if self.accident_detected else '❌ 否'}") + print(f" - 平均处理帧率 → {avg_fps} FPS") + + +# 供外部导入 +__all__ = ["AccidentDetector"]