Greyscale Image Converter in Python

Reading time ~1 minute

Back in CPSC 101, we did a lot of image editing. We only used PPM images in class and I was thinking the other day about how even though I hated the project at the time, it was pretty interesting. So, I decided to re-code some of it in Python. Basically, all I’ve done is written a script that will allow the conversion of any color PPM image into greyscale. I feel like this can be pretty handy, as it’s fast and seems to do a pretty good job.

Here’s the before image (Ironically, I had to convert to another image format before posting):

alt text

And here’s the output:

alt text

A few particulars about the files and scripts:

  • I used the P3 header, which is easier because it allows you just to write ASCII into the file.

  • Also, you must ensure that the header of the input photo must only have 3 lines. I’ve found that if you convert a JPEG from Gimp, it adds its on line in the header. It was easy to just do ‘nano in.ppm’ in terminal and remove the line they put in.

  • The script can be run using ‘python <input filename>’

  • This ONLY works on PPM images. Maybe someday I’ll look into another image format.

And that’s it!

Here’s the code, check it out:

#Greyscale Converter
#Spencer Smith
#Mar. 22, 2010
import sys
#Open file
FILE = open(sys.argv[1],"r")
#Read input file
indata = FILE.readlines()
#Strip header
del indata[0]
size = indata[0]
del indata[0:2]
#Open output image
GREY = open("out_grey.ppm","w")
#Write header to output image
#Calculate and write RGB values
counter = 0
for x in indata:
  val = x
  if counter % 3 == 0:
    val1 = int(val)*.3
  elif counter % 3 == 1:
    val2 = int(val)*.59
  elif counter % 3 == 2:
    val3 = int(val)*.11
  px = val1+val2+val3
  GREY.write("%d\n%d\n%d\n" % (int(px),int(px),int(px)))
  counter = counter + 1
#Close files

KubeDNS Tweaks for Performance

Hey y'all. Wanted to document some of the stranger bits I've encountered while running Kubernetes with one of my clients. We've finally g...… Continue reading

Reporting to CloudWatch from CoreOS

Published on March 04, 2017