Commit Graph

6 Commits

Author SHA1 Message Date
wwh
7aa71c5f83 feat: 新增人员徘徊/静止行为分析功能
本次提交实现了完整的人员行为分析系统,包括:
1. 新增基于位置和跟踪ID的两种行为检测算法
2. 新增徘徊检测服务与行为处理器模块
3. 前后端集成算法配置界面与告警展示
4. 支持图片和视频流场景下的行为分析
5. 新增算法配置接口与文档说明

具体改动:
- 新增loitering_detection模型目录与算法实现
- 新增AlgorithmConfig组件实现可视化配置
- 扩展图片/视频检测接口支持算法参数传递
- 新增行为告警推送与前端展示页面
- 优化检测服务,集成行为分析逻辑
- 移除冗余日志输出,完善代码注释
2026-05-19 09:17:09 +08:00
wwh
2691761f01 refactor: 整理项目配置与代码,更新文档
1.  将turbo.json中的pipeline重命名为tasks
2.  修复Vue模板中小于号转义问题,清理冗余代码与样式
3.  更新README文档,补充后端启动步骤,移除旧贡献指南
4.  更新前端包的rollup依赖版本与包名
2026-05-18 15:34:20 +08:00
wwh
1813bc8c0b refactor: 重构图片检测功能,拆分组件并优化返回格式
1. 后端api修改:将图片结果返回从文件路径改为base64格式,移除本地文件存储
2. 新增图片检测组件ImageDetection.vue,封装独立的图片检测UI逻辑
3. 重构Home页面,使用tab切换图片/视频检测模块,简化原有布局
4. 更新web包名与rollup依赖版本
2026-05-18 15:15:29 +08:00
wwh
364b4ac189 feat: Add AI model files with Git LFS
Add 5 YOLO model files for video detection:
- fire_detection/best.pt (61MB) - YOLOv10 fire and smoke detection
- helmet_detection/yolov8n.pt (6.2MB) - YOLOv8 helmet detection
- crowd_detection/yolov8l.pt (84MB) - YOLOv8 crowd gathering detection
- smoking_detection/smoking_yolov8n.pt (5.9MB) - YOLOv8 smoking detection
- loitering_detection/yolov8n.pt (6.2MB) - YOLOv8 loitering detection

Configure Git LFS to track *.pt files
Update .gitignore to allow .pt files (now managed by LFS)
2026-05-18 11:21:22 +08:00
wwh
54ac754660 docs: Add model setup documentation and script
- Add models/README.md with model file information
- Add scripts/setup-models.sh for easy model linking
- Document how to obtain and setup model files
2026-05-18 10:56:40 +08:00
wwh
8fb58c75fe Initial commit: Video detection platform with YOLO models
Features:
- Fire detection (YOLOv10)
- Helmet detection (YOLOv8)
- Crowd detection (YOLOv8)
- Smoking detection (YOLOv8)
- Loitering detection (YOLOv8)

Tech Stack:
- Frontend: Vue 3 + Vite + Element Plus
- Backend: FastAPI + WebSocket
- Monorepo: pnpm workspace + Turbo
- Docker support included
2026-05-18 10:54:10 +08:00