burn

  1. downloadJetson download center | NVIDIA DeveloperAnd extract the image filesd-blob.img

⚠️ Pay attention to the version of your board here

  • You need to install the 4GB version
  • The new board says the previous generation has excess performance and needs to install a 2GB versionJetson Nano 2GB Developer Kit

(Burned the 4GB version twice in the new board and it never lit up…)

  1. Burn to a microSD card

It is recommended to use the official recommendation balenaEtcher – Flash OS images to the specific flow of SD CARDS and USB drives reference: Jetson Nano Developer suite portal | NVIDIA Developer

  1. Connect peripherals (mouse, keyboard, monitor) and start

The power supply

If the NVIDIA interface keeps appearing/disappearing/appearing/disappearing during the boot process, it may be because the machine failed in the middle of boot and then restarted, which involves the power supply problem:

  • If DP power is used, install jumper caps
  • Make sure the wattage is up to the requirement
  • Lines and heads should also conform to specifications

Novice’s Manual (1) : Install DC power for Jetson Nano

The camera

Rookie’s Manual (2) : Install a Raspberry PI camera on the Jetson Nano

GST - launch - 1.0 nvarguscamerasrc!'video/x-raw(memory:NVMM),width=3820, height=2464, framerate=21/1, format=NV12' ! nvvidconv flip-method=0 ! 'video/x-raw,width=960, height=616' ! nvvidconv ! nvegltransform ! nveglglessink -e
Copy the code

GitHub – JetsonHacksNano/ cs-camera: Simple example of using a CSI-Camera (like the Raspberry Pi Version 2 camera) with the NVIDIA Jetson Nano Developer Kit

python simple_camera.py
python face_detection.py
Copy the code

Software environment

  • [[Switching Python versions]]
  • Python: Jetson Nano installation and configuration _ Deep Learning -CSDN blog

Note that it says to change a piece of code, do not change ❌

  • NVCC, OpencV, cuDNN environment: Note (5) Jetson Nano basic environment configuration _SWorld-CSDN blog
  • Turn off the GUI environment: Jetson Nano only has 4G memory, the GUI will take up a large part of the memory (800M)
sudo init 3     # Close desktop
sudo init 5     # Restart the desktop
Copy the code

Download the software

⚠️ Note that Nano is arm64 architecture, select the version clearly when installing software

Install the deb file
sudo dpkg -i xxx.deb
Copy the code

Resources

  • Official resources download: Jetson Nano Developer suite portal | NVIDIA Developer
  • Nvidia Jetson Nano introduction and User guide – Zhihu
  • Jetson Nano starts from scratch (2) : Hardware