Re: [ft-devel] GSoC - Image comparison in test framework

[Arvinder Bhathal]

I've submitted a draft proposal and I appreciate any comments on it. I
also submitted it as a final proposal just to be safe this close to
the deadline but I can edit my final submission anytime.

> Indeed, MD5 will not tell how visible are the differences. AFAIK, 64-bit and
> 32-bit FreeType will produce different results today, but those differences
> would not be visible. I would appreciate some quantitative measure how
> visible are the detected differences. Font rendering quality is mostly about
> contrast, so some integral contrast measure and how it changed would also be
> nice.

I elaborated a bit in the proposal but I went with the root mean
square measurement for contrast. Given the appearance of font glyphs,
it made the most sense to take into account the average intensity of
the glyph and the intensity of each pixel to obtain a nice overall
measurement of contrast. In other image contrast calculations that
only take into account the most intense and least intense pixel, we
might find the contrast to be the same for two similar (but different
on a pixel level) glyphs.