┌─────────────┐     ┌──────────────────────┐     ┌───────────┐
│   Frontend  │────▶│   Backend (FastAPI)  │────▶│  YOLO模型  │
│  (静态页面)  │◀────│   REST API 服务      │◀────│  人脸检测   │
└─────────────┘     └──────────────────────┘     └───────────┘
                            │
                            ▼
                     ┌───────────┐
                     │  SQLite   │
                     │ 标注存储   │
                     └───────────┘

yolo_face_boxes/
├── backend/
│   ├── main.py            # FastAPI 入口，路由注册
│   ├── detector.py        # YOLO 检测器封装（ultralytics）
│   ├── database.py        # SQLite 数据库操作
│   ├── schemas.py         # Pydantic 数据模型
│   └── requirements.txt   # Python 依赖
├── frontend/
│   ├── index.html         # 主页面
│   ├── css/style.css      # 样式
│   └── js/app.js          # 交互逻辑（图片展示+标注）
├── data/
│   ├── uploads/           # 上传图片存储
│   └── db/                # SQLite 数据库文件
└── README.md

1. 用户上传教室合照
      ↓
2. 后端 YOLO 检测，返回人脸框坐标
      ↓
3. 前端绘制图片 + 人脸框
      ↓
4. 用户逐个点击人脸框，输入姓名/学号
      ↓
5. 提交标注 → 存入数据库
      ↓
6. 生成/导出签到表

┌─────────────────────────────────────────────────┐
│               前端 SPA (纯静态)                  │
│  ┌──────────┐ ┌─────────────┐ ┌──────────────┐  │
│  │ 图片上传  │ │ Canvas标注   │ │ localStorage │  │
│  │ XLSX导入  │ │ 绘制/删除框  │ │ 键值对持久化  │  │
│  └─────┬────┘ └──────┬──────┘ └──────┬───────┘  │
│        └──────────────┼───────────────┘          │
│              XLSX导出签到表                       │
└──────────────────────┬──────────────────────────┘
                       │ HTTP (仅检测时调用)
┌──────────────────────▼──────────────────────────┐
│            后端 Flask API (无状态)                │
│   POST /api/detect — YOLO推理 (CPU/ONNX)         │
│   GET  /api/health — 健康检查                    │
└─────────────────────────────────────────────────┘

backend/
  app.py              # Flask API (YOLO检测)
  requirements.txt    # Python依赖
frontend/
  index.html          # 主页面
  css/style.css       # 样式
  js/storage.js       # localStorage键值对模块
  js/canvas.js        # Canvas绘制/删除/标注模块
  js/api.js           # 后端API调用模块
  js/app.js           # 主应用逻辑
  lib/xlsx.full.min.js # SheetJS XLSX库
Dockerfile            # CPU部署容器

$:
  vscode:
    - docker:
        build: .ide/Dockerfile
      services:
        - vscode
        - docker
      stages:
        - name: install dependencies
          script: pip install -r requirements.txt
        - name: download model weights
          script: |
            mkdir -p app/weights
            if [ ! -f app/weights/yolov8n-face.pt ]; then
              curl -L -o app/weights/yolov8n-face.pt https://github.com/lindevs/yolov8-face/releases/download/1.0.1/yolov8n-face-lindevs.pt
            fi

docker login docker.cnb.cool -u cnb -p ${CNB_TOKEN}
docker build -t ${CNB_DOCKER_REGISTRY}/${CNB_REPO_SLUG_LOWERCASE}:latest .
docker push ${CNB_DOCKER_REGISTRY}/${CNB_REPO_SLUG_LOWERCASE}:latest

$:
  vscode:
    - docker:
        build: .ide/Dockerfile   # 每次启动时构建，耗时较长

$:
  vscode:
    - docker:
        image: ${CNB_DOCKER_REGISTRY}/${CNB_REPO_SLUG_LOWERCASE}:latest  # 直接拉取，秒级启动

 > [6/9] RUN pip install --no-cache-dir -r requirements.txt:

355.6    ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 536.2/536.2 kB 6.9 MB/s eta 0:00:00

355.6 Downloading six-1.17.0-py2.py3-none-any.whl (11 kB)

356.0 ERROR: THESE PACKAGES DO NOT MATCH THE HASHES FROM THE REQUIREMENTS FILE. If you have updated the package versions, please update the hashes. Otherwise, examine the package contents carefully; someone may have tampered with them.

356.0     unknown package:

356.0         Expected sha256 7aa2f9bbc6d4595ba72138026b2074be1233186150e9292865e04b7a63b8c67a

356.0              Got        726f4dea5e6fb9c7aecc9360ea28a2efb2d11843462f0000d175e284fa9e27fc

356.0 

356.3 

356.3 [notice] A new release of pip is available: 24.0 -> 26.0.1

356.3 [notice] To update, run: pip install --upgrade pip

------

Dockerfile:27

--------------------

  25 |     # Pre-install Python dependencies

  26 |     COPY requirements.txt .

  27 | >>> RUN pip install --no-cache-dir -r requirements.txt

  28 |     

  29 |     # Download yolov8n-face model weights

--------------------

ERROR: failed to build: failed to solve: process "/bin/sh -c pip install --no-cache-dir -r requirements.txt" did not complete successfully: exit code: 1

Finished, code: 1, duration: 7m 11s

- RUN pip install --no-cache-dir -r requirements.txt
+ RUN pip install --upgrade pip && pip install --no-cache-dir -r requirements.txt