Index of /teaching/2021/L314/assignment4b

	Name	Last modified	Size	Description
	Parent Directory		-
	README.txt	2019-12-11 16:51	1.0K
	raw/	2021-01-15 15:42	-
	test1.png	2019-12-11 16:51	5.0M
	test1c.png	2019-12-11 16:51	7.8M
	test2.png	2019-12-11 16:51	3.7M
	test2c.png	2019-12-11 16:51	11M
	test3.png	2019-12-11 16:51	4.5M
	test3c.png	2019-12-11 16:51	18M

JPEG compression artefact data set
----------------------------------

Markus Kuhn -- 2019-12-11

I took three 3024 x 4032 raw photographs (camera: Google Pixel 3), and
converted the resulting *.dng files into six *.ppm files, using the
ufraw software (using six different brightness adjustments, exporting
each photo once in colour and once as a grayscale image.)

I then passed each photo through one of the following five processing
pipelines:

  pnmtopng
  cjpeg -grayscale -quality $Q1 | djpeg | pnmtopng >$@
  cjpeg -grayscale -quality $Q2 | djpeg | cjpeg -grayscale -quality $Q3 | djpeg | pnmtopng
  cjpeg -quality $Q4 | djpeg | pnmtopng
  cjpeg -quality $Q5 | djpeg | cjpeg -quality $Q6 | djpeg | pnmtopng

In other words, the image data in each of the files

  test1.png
  test1c.png
  test2.png
  test2c.png
  test3.png
  test3c.png

has gone either zero, one or two times through a JPEG
compression/decompression step.

How often did each image go through a JPEG codec?