temporal aliasing

[gusl]

During the protest against the closure of the Health Contact Center, a bus stop mysteriously displays a message in Nordic (is it Norwegian?).

In photographing it, I noticed that I had to control the shutter speed manually.




Is it possible to figure out the frequency of the lights from this?

An interesting skill would be to time your own saccades, and to count the number of flashes you see in between. Maybe I could pick this up with a couple of hours' training.

Comments

[bhudson.livejournal.com]

I'm going to bet 60 Hz, given that the 80-Hz image shows some flicker still (some pixels are darker). And because mains in north America are at 60 Hz.

[gustavolacerda.livejournal.com]

I think we should push for ISO settings to be more, er, irrational!

[gustavolacerda.livejournal.com]

is ISO measured in Hz? Just making sure.

So a shot at ISO 80 is open for 1/80s?

This is surprising, since AFAIR you're supposed to use higher ISOs in dark places.

[bhudson.livejournal.com]

Shutter speed is measured in fractions of seconds, or in inverse seconds, often both depending on scale.

ISO is the film speed -- how sensitive the detector is. Higher numbers indicate greater sensitivity (so fewer photons are required to trigger the detector).

[jcreed.livejournal.com]

google says it's "please see timetable" as norwegian

[psifenix.livejournal.com]

In any case, 'Nordic' is preferable to 'Scandinavian,' as it clearly is not Finnish.

[gustavolacerda.livejournal.com]

I agree with your recommendation. However, Finland is considered Nordic too.

[stepleton.livejournal.com]

The term Nordic appears to be more specific within the context of languages.

[widdertwin.livejournal.com]

You've got that reversed. Nordic refers to the geographical region, which includes Finland. Scandinavian refers only to Norway, Denmark, and Sweden.

The language is either Danish or Norwegian. Probably Norwegian, though. I think Danish would use the preposition "på"; "venligst se på tidtabellen". It's been a while since I've used either language, though.

[gustavolacerda.livejournal.com]

Interesting. In my head, "på" = "by" (EN). "Please see by timetable"?

[widdertwin.livejournal.com]

Oops, I accidentally said "tidatabellen". That's not a thing at all. "På" means a number of things depending on the context, but is most frequently translated as "at: or "on". Just out of curiosity, what made you think "by"? Is this something Dutch that I'm unaware of?

[gustavolacerda.livejournal.com]

I have no idea why I thought that, but might be related to the little Swedish I learned when I was bored in Summer 1999.

It sounds like "på" is "op"(NL) :-) which usually means "on" or "at".

[gustavolacerda.livejournal.com]

It's obvious to me that "tidtabellen" is "the timetable". In Dutch, "time" = "tijd". But the "Vennligst" stumped me.

I imagine this was triggered by someone who works for Transit noticing the protest.

Btw, what are the units for ISO? 80 what?

[peamasii.livejournal.com]

so aliasing is a mechanism to improve clarity and reduce smearing, by saccade masking? interesting.

[gustavolacerda.livejournal.com]

Aliasing is a phenomenon in signal processing that happens whenever the sampling rate of the device is below the Nyquist frequency. It's what you try to avoid.

By "saccade" I mean the movement of the human eye.

[peamasii.livejournal.com]

The brain might be shuttering the visual information by saccades to avoid image smearing and to improve visual recognition. So you see a moving image, the saccades break it up into separate stills, and the brain shutters out certain frames to allow you to see clearly only a certain amount of frames per second. Is that how it might work?

[gustavolacerda.livejournal.com]

whether the human visual system has a fixed sampling rate is an interesting question... I would bet this is true to some extent, and that in fact we see some temporal aliasing. (Likewise, maybe pupil dilation corresponds to ISO settings)

It would only be logical for saccades to correlate with snapshots, so I bet it's true. Your smearing prevention theory makes sense to me, and I imagine there should be a sharp drop in shape recognition when something moves faster than a certain threshold.

But somehow I suspect that our visual system can't be that similar to camcorders! :-)

[peamasii.livejournal.com]

Hey, it can't be that difficult for evolution to simulate a camcorder... I'd sooner believe that the aliasing is made neurologically and that physically the analogous cortex-retina system is working in a flux. Maybe you can ask a neurologist next time you meet one. Apparently movies use 24fps because at that rate motion can be perceived smoothly while any higher framerate would not provide any useful additional visual information?

I like your idea that we might be able to develop the skill of perceiving 'speed' by the differential between saccade snapshots (that would presume some control over a mostly unconscious activity though).