skye.yue
ea618fd674
- Updated argument parsing in manage_api.py to include new threshold parameters. - Enhanced _config_payload to include thresholds and webhook configurations. - Modified _build_summary to track queue metrics and adjust alert reporting. - Refactored DwellEngine to utilize queue thresholds for alerting and reporting. - Added queue metrics calculations and status change tracking in dwell_engine.py. - Updated notifier.py to support posting JSON events to webhooks. - Adjusted example configuration to reflect new threshold parameters. - Enhanced Docker entrypoint script for better process management. - Updated tests to cover new queue metrics and thresholds. - Improved ManagedServiceDetail and ManagedServices Vue components to display queue metrics.
Store Dwell Alert on 192.168.5.108
Local planning project for a single-camera RTSP long-stay alert system running on 192.168.5.108.
Documents:
docs/plans/2026-04-15-store-dwell-alert-design.mddocs/plans/2026-04-15-store-dwell-alert.md
Scope:
- Single RTSP store camera
YOLO11nperson detection- Full-frame tracking and lightweight re-association
- Staff whitelist exclusion
- Local JSONL event sink
- Half-hour dwell reports
- Offline bundle packaging for
.108-class Ubuntu hosts - Chinese operations guide:
README_zh.md
Runtime:
- Default config path:
config/108.local.yaml - Event sink path:
logs/events.jsonl - Continuous mode:
python -m app.main - Single-frame smoke test:
python -m app.main --once - Management API mode:
python -m app.manage_api --config config/local.yaml --port 18081
Staff gallery:
- Directory:
data/staff_gallery/ - Supported formats:
*.jsonsignature files*.jpg,*.jpeg,*.png,*.bmp,*.webpsample crops
- JSON example:
{
"staff_id": "staff_001",
"signature": [0.182, 0.244, 0.317]
}
Offline bundle:
- Target host assumption: Ubuntu 24.04, Python 3.12, NVIDIA GPU compatible with the
.108runtime stack - Bundle script:
bash scripts/package_bundle.sh - Bundle output:
dist/store_dwell_alert_bundle_<date>.tar.gz - Bundle prerequisites before packaging:
- place Linux wheels under
wheelhouse/ - place
yolo11n.ptunderweights/yolo11n.pt
- place Linux wheels under
Offline install on target host:
- Extract
store_dwell_alert_bundle_<date>.tar.gz - Edit
scripts/run.shand setRTSP_URL - Run
sudo bash scripts/install.sh - The installer enables and starts
store-dwell-alert.service
Optional service install:
- Run
bash scripts/install_service.sh - Start with
sudo systemctl start store-dwell-alert.service
Docker image:
- Required runtime weight in build context:
weights/yolo11n.pt - Build:
docker build -t store-dwell-alert:test . - Compose:
docker compose up --build store-dwell-alert - Health:
http://127.0.0.1:18081/api/manage/health - Default container entrypoint seeds
config/local.yamlfromconfig/config.example.yamland then starts the management API - Persisted host paths:
./config -> /app/config./data -> /app/data./logs -> /app/logs
中文离线部署说明
适用环境：
Ubuntu 24.04Python 3.12- 已安装 NVIDIA 驱动，
nvidia-smi可用 - 已安装
ffmpeg
离线包准备：
- 成品包示例：
store_dwell_alert_bundle_2026-04-16.tar.gz - 包内已经包含：
- 项目源码
wheelhouse/离线依赖weights/yolo11n.ptscripts/install.shscripts/run.shscripts/install_service.sh
部署步骤：
- 把离线包传到目标机器
- 解压：
tar -xzf store_dwell_alert_bundle_2026-04-16.tar.gz
cd store_dwell_alert_bundle
- 执行离线安装：
bash scripts/install.sh
- 编辑运行脚本，至少改这几个变量：
RTSP_URLCAMERA_IDEVENT_SINK_PATH
vim scripts/run.sh
- 手工启动：
bash scripts/run.sh
常见输出位置：
- 事件结果：
logs/events.jsonl - 本地配置：
config/local.yaml - 店员图库：
data/staff_gallery/
如果需要做店员排除：
- 把店员样本图放到
data/staff_gallery/ - 或者放入对应的
*.json特征文件
配置完成后的常驻运行：
- 安装服务：
bash scripts/install_service.sh
- 启动服务：
sudo systemctl start store-dwell-alert.service
- 查看状态：
sudo systemctl status store-dwell-alert.service
- 停止服务：
sudo systemctl stop store-dwell-alert.service
- 开机自启：
sudo systemctl enable store-dwell-alert.service
排查建议：
scripts/install.sh失败时，先确认python3.12、ffmpeg、nvidia-smi都存在scripts/run.sh失败时，先确认RTSP_URL已改成真实地址- 如果程序启动了但没有结果，先看
logs/events.jsonl是否生成 - 如果识别不到店员，先检查
data/staff_gallery/是否有有效样本