In this github, I will show you how to run the Hardhat Detectation with Jetson Nano.
The whole tutorial of hard hat detectation can be learn from this link: https://wiki.seeedstudio.com/HardHat
Here are 2 ways to run this demo.
Method 1: (for those who want to run the demo directally) Prepare a SD card larger than 16 GB and flash the image file into your device
Method 2: (for those who donnot want to waste time to download image and have access to google.com) Use command line to install dependencies
All the methods can let you fully experience Edge Impulse if you want to use your own account and project.
This is the system image file for Jetson nano created by SeeedStudio with necessary environment.
Everyone can download this file which is for personal study use only.
Click here to download the image file: https://drive.google.com/file/d/13rl93ZB3LOu0jbArKETDJ9V-N7MJw0uK/view?usp=sharing
username:jetson-nano
password:nvidia
python3 hard_hat_detectation.py /home/seeedstudio/modelfile.eim
sudo apt-get update
sudo apt install software-properties-common
sudo add-apt-repository ppa:deadsnakes/ppa
sudo apt install python3.7
Update python 3 point to python3.7
sudo update-alternatives --install /usr/bin/python3 python3 /usr/bin/python3.6 1
sudo update-alternatives --install /usr/bin/python3 python3 /usr/bin/python3.7 2
sudo update-alternatives --config python3
Press Enter and choose 2
python3 -m pip install --upgrade pip
sudo pip3 install scikit-build
sudo pip3 install cython
Please follow this official link: https://docs.edgeimpulse.com/docs/linux-python-sdk
Step 5. (Optional for those who want to use their own project and model) Deploy your Jetson to your project
Please follow this official link: https://docs.edgeimpulse.com/docs/nvidia-jetson-nano
sudo apt install toilet
Step 7. Download my modefile and python demo (modelfile.eim
hardhat_detectation.py
device_patches.py
) and then run it with:
python3 hard_hat_detectation.py /home/<your_device_name>/modelfile.eim
PS: If you donnot want to use my model, you can download your own model to the location: /home/<your_device_name>
edge-impulse-linux-runner --download modelfile.eim
then repeat the command python3 hard_hat_detectation.py /home/<your_device_name>/modelfile.eim