Skip to content

ZhenWZ/NICT2021RTT_Group17

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

20 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

门禁监测助手 / Mask Detection Assistant

Version Date Description Other
v1.0.0 2021/07/22 口罩监测模型部署到 ART-Pi 中

1. 项目说明 / Introduction

该项目将口罩监测模型部署到RT-Thread提供的ART-Pi开发板并通过gc0328读取图像数据,最后通过多媒体扩展版输出结果。

In this project, We trained a mask detection model and deployed it onto the ART-Pi development boarded. The development board takes in images collected by the gc0328 camera module, and output the result of mask detection model with an intergrated screen.

2. 环境 / Environment

  • 操作系统: Windows10

  • X-CUBE-AI: 5.2.0

  • RT-Thread Studio: 2.1.1

  • RT-AK

    RT-AK仓库

3. 元件 / Components

  • ART-Pi
  • ART-Pi 多媒体扩展版
  • gc0328

4. 步骤 / Steps

4.0 准备文件

  • RT-Thread工程
  • RT-AK
  • 准备好的模型
  • X-CUBE-AI

4.1 部署模型 / Deployment

在命令行中进入到RT-AK\rt_ai_tools输入

python aitools.py --project="工程路径" --model="模型路径" --model_name="mask" --platform=stm32 --ext_tools="X-CUBE-AI路径" --clear

例如:

D:\RT-ThreadAI\RT-AK\rt_ai_tools>python aitools.py --project=D:\RT-ThreadAI\mask_detect --model=D:\RT-ThreadAI\RT-AK\rt_ai_tools\Models\mask_detect.tflite --model_name=mask --platform=stm32 --ext_tools=D:\RT-ThreadAI\stm32ai-windows-5.2.0\windows --clear

5. 训练模型 / Training

5.1 环境 / Environment

  • tensorflow==2.4.0
  • opencv-python==4.5.1
  • matplotlib==3.3.3

5.2 数据集 / Datasets

5.3 网络结构 / Model Structure

  • 两层卷积 + 一层全连接

model

5.4 训练模型 & 验证模型 / Traning & Validating

  • 训练模型 / Training

training

  • 验证训练模型的准确率 / Accuracy

evaluate

6. 运行结果 / Results

感谢名单 / Acknowledgements

  • Derekduke
  • lebhoryi

About

Mask Detection based on Art-Pi

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published