Raspberry Pi SH1106
Neil Haddley • July 24, 2023
OLED display
The Raspberry Pi can output text (and simple graphics) to an OLED display using Python.
OLED wiring
i2cdetect -y 1
Luma-OLED Driver
I installed the Luma-OLED driver by following these instructions:
https://luma-oled.readthedocs.io/en/latest/install.html
I installed the driver with this command
$ sudo -H pip3 install --upgrade luma.oled
I enabled I2C communications using the raspi-config application
$ sudo raspi-config
Interface Options | I2C | enable ARM I2C
i2c permissions
I followed the steps described by Alexander Rüedlinger to allow non-root users to access i2c devices (specifically the OLED display).
1) Create new user group called i2c:
$ sudo groupadd i2c
2) Change the group ownership of /dev/i2c-1 to i2c:
$ sudo chown :i2c /dev/i2c-1
3) Change the file permissions of the device /dev/i2c-1 so users of the i2c group can read and write to the device:
$ sudo chmod g+rw /dev/i2c-1
4) Add your user to the group i2c:
$ sudo usermod -aG i2c <username>
Python code
The hello.py code running on configured hardware
Displaying an image on the OLED display
I downloaded a stop sign image and updated the hello.py code to display the image on the OLED display.
Stop sign
Using OpenCV and luma.oled to display video on the OLED display
OpenCV (Open Source Computer Vision Library) is an open-source computer vision and machine learning software library. It provides a comprehensive set of tools and functions that enable developers to perform various computer vision tasks, such as image and video processing, object detection, facial recognition, and feature extraction.
I used the cv2 library to read frames from a dog video I downloaded from:
https://pixabay.com/videos/puppy-dog-playful-beach-sand-play-4740/
Then I updated the code to send image captured by the webcam to the OLED display.
Dog video
Webcam
hello.py
TEXT
1from luma.core.interface.serial import i2c, spi, pcf8574 2from luma.core.interface.parallel import bitbang_6800 3from luma.core.render import canvas 4from luma.oled.device import ssd1306, ssd1309, ssd1325, ssd1331, sh1106, sh1107, ws0010 5from time import sleep 6 7# rev.1 users set port=0 8# substitute spi(device=0, port=0) below if using that interface 9# substitute bitbang_6800(RS=7, E=8, PINS=[25,24,23,27]) below if using that interface 10serial = i2c(port=1, address=0x3C) 11 12# substitute ssd1331(...) or sh1106(...) below if using that device 13device = sh1106(serial) 14 15with canvas(device) as draw: 16 draw.rectangle(device.bounding_box, outline="white", fill="black") 17 draw.text((30, 40), "Hello World", fill="white") 18 19sleep(10)
stop_sign_display.py
TEXT
1from luma.core.interface.serial import i2c, spi, pcf8574 2from luma.core.interface.parallel import bitbang_6800 3from luma.core.render import canvas 4from luma.oled.device import ssd1306, ssd1309, ssd1325, ssd1331, sh1106, sh1107 5from time import sleep 6 7from PIL import Image 8 9serial = i2c(port=1, address=0x3C) 10device = sh1106(serial) 11 12width = device.width 13height = device.height 14 15image = Image.open("stop.png") 16 17image_r = image.resize((width,height), Image.LANCZOS) 18# Image.BICUBIC is another option 19image_bw = image_r.convert("1") 20 21device.display(image_bw) 22 23sleep(10)
dog_display_window_and_oled.py
TEXT
1from luma.core.interface.serial import i2c, spi, pcf8574 2from luma.core.interface.parallel import bitbang_6800 3from luma.core.render import canvas 4from luma.oled.device import ssd1306, ssd1309, ssd1325, ssd1331, sh1106, sh1107 5 6from time import sleep 7 8import cv2 9 10# OpenCV follows BGR color convention and PIL follows RGB color convention 11 12# Python Imaging Library 13from PIL import Image 14 15serial = i2c(port=1, address=0x3C) 16device = sh1106(serial) 17 18width = device.width 19height = device.height 20 21cap = cv2.VideoCapture("dog.mp4") 22while True: 23 ret, frame = cap.read() 24 if not ret: 25 break 26 cv2.imshow("Img", frame) 27 28 color_coverted = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB) 29 pil_image = Image.fromarray(color_coverted) 30 image_r = pil_image.resize((width,height), Image.LANCZOS) 31 # Image.BICUBIC is another option 32 image_bw = image_r.convert("1") 33 34 device.display(image_bw) 35 36 key = cv2.waitKey(1) 37 if key == 27: 38 break 39 40cap.release() 41cv2.destroyAllWindows()
webcam_display_window_and_oled.py
TEXT
1... 2cap = cv2.VideoCapture(0) 3...