This is the fourth entry in my 8 part series on adding some object detection to my Raspberry Pi.
In this section I am going to cover the “other” parts of the Pi installation that I need. This will include some preferences as well as getting python up and running.
First, I am going to primarily develop on the Pi using VNC. In my last post, I talked about getting VNC access. While it works out of the box it is less than ideal when your screen resolution is the size of a note card. To change the default resolution:
sudo raspi-configto have the Pi settings come up
- Navigate to **Advanced Options**
From here you can change to whatever you like. I did 1024×720 since that is plenty big. Just know that the higher the resolution the more the Pi has to work.
While python is installed I will need to add some updates. I am using the guide from Raspberry Pi here. After running
sudo apt update I am able to download the Python3 camera library using
sudo apt install python3-picamera. Based on your install of the OS this might already be ready.
RaspberryPi comes with Thonny for an IDE if you want to go that route. I am undecided how I will go about doing all of this work. It will probably be a combination of local development, Thonny on the Pi, and some command line work.
I did have to install TensorFlow v2. To do that I went into the Tools menu of Thonny and searching PyPi for ‘Tensorflow’. It found v2.1.0 and as I tried to install it there was a message about only seeing versions 1.14 and below. This is because of an issue with the ARM processor that is running on the Pi. I explain in part 5 in more detail but I was able to find an ARM built whl.
This was another short post but there were a few additions that I wanted to add to my previous post that didn’t really fit in.
My next post will be setting up the camera. This will include hooking up the hardware as well as writing a little bit of test code.