Skip to content

Instantly share code, notes, and snippets.

@forslund
Last active November 17, 2017 15:18
Show Gist options
  • Save forslund/9d8805fd7adb9e74ec3ea321e1676a21 to your computer and use it in GitHub Desktop.
Save forslund/9d8805fd7adb9e74ec3ea321e1676a21 to your computer and use it in GitHub Desktop.
Simple program to convert png to mycroft display images
import sys
from PIL import Image
def mouth_display_png(image_absolute_path, threshold=70,
invert=False, x=0, y=0):
"""Converts a png image into the appropriate encoding for the
Arduino Mark I enclosure.
NOTE: extract this out of api.py when re structuing the
enclosure folder
Args:
image_absolute_path (string): The absolute path of the image
threshold (int): The value ranges from 0 to 255. The pixel will
draw on the faceplate it the value is below a
threshold
invert (bool): inverts the image being drawn.
x (int): x offset for image
y (int): y offset for image
"""
# to understand how this funtion works you need to understand how the
# Mark I arduino proprietary encoding works to display to the faceplate
img = Image.open(image_absolute_path).convert("RGBA")
img2 = Image.new('RGBA', img.size, (255, 255, 255))
width = img.size[0]
height = img.size[1]
# strips out alpha value and blends it with the RGB values
img = Image.alpha_composite(img2, img)
img = img.convert("L")
# crop image to only allow a max width of 16
if width > 32:
img = img.crop((0, 0, 32, height))
width = img.size[0]
height = img.size[1]
# crop the image to limit the max height of 8
if height > 8:
img = img.crop((0, 0, width, 8))
width = img.size[0]
height = img.size[1]
encode = ""
# Each char value represents a width number starting with B=1
# then increment 1 for the next. ie C=2
width_codes = ['B', 'C', 'D', 'E', 'F', 'G', 'H', 'I', 'J', 'K', 'L',
'M', 'N', 'O', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W',
'X', 'Y', 'Z', '[', '\\', ']', '^', '_', '`', 'a']
height_codes = ['B', 'C', 'D', 'E', 'F', 'G', 'H', 'I']
encode += width_codes[width - 1]
encode += height_codes[height - 1]
# Turn the image pixels into binary values 1's and 0's
# the Mark I face plate encoding uses binary values to
# binary_values returns a list of 1's and 0s'. ie ['1', '1', '0', ...]
binary_values = []
for i in range(width):
for j in range(height):
if img.getpixel((i, j)) < threshold:
if invert is False:
binary_values.append('1')
else:
binary_values.append('0')
else:
if invert is False:
binary_values.append('0')
else:
binary_values.append('1')
# these values are used to determine how binary values
# needs to be grouped together
number_of_top_pixel = 0
number_of_bottom_pixel = 0
if height > 4:
number_of_top_pixel = 4
number_of_bottom_pixel = height - 4
else:
number_of_top_pixel = height
# this loop will group together the individual binary values
# ie. binary_list = ['1111', '001', '0101', '100']
binary_list = []
binary_code = ''
increment = 0
alternate = False
for val in binary_values:
binary_code += val
increment += 1
if increment == number_of_top_pixel and alternate is False:
# binary code is reversed for encoding
binary_list.append(binary_code[::-1])
increment = 0
binary_code = ''
alternate = True
elif increment == number_of_bottom_pixel and alternate is True:
binary_list.append(binary_code[::-1])
increment = 0
binary_code = ''
alternate = False
# Code to let the Makrk I arduino know where to place the
# pixels on the faceplate
pixel_codes = ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H',
'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P']
for binary_values in binary_list:
number = int(binary_values, 2)
pixel_code = pixel_codes[number]
encode += pixel_code
return encode
if __name__ == '__main__':
print mouth_display_png(sys.argv[1])
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment