Frage von colacao:
Hello,
for an experiment, I have measured the Pupillenvergroesserung with people while they watched a video. The values were later issued as a curve.
Now the idea that the light in the whole video is to be measured to subtract these values then values of the pupils to make a meaningful analysis.
It has already spoken about light metering (http://forum.slashcam.de/fachwort-fur-ein-gerat-um-light-zu-messen-vt31100.html?highlight=lichtmesser), but I still have a Problem:
As the light is measured over an entire video? Specifies the device from one value per frame? Should we do that by hand for each frame separately, or a device can automatically record everything?
How can I make these values in relation to Pupillenvergroesserung?
Unfortunately I am still rather clueless in the area and I would be happy about any nice answer. Or maybe someone knows of a specialist who could help me. Gladly synonymous in the private e-mail.
Thank you.
Antwort von Stefan:
Spontaneously I would have suggested Avisynth and a plugin / filter. The filter would have to calculate the average brightness of a frame and write, for example in a text table. Which you can then in Excel ö.ä. at the desired time and resources vernudeln with the measurement results.
How did you do the measurements of the pupillary opening? Perhaps there are videos available that you could evaluate with respect to the amount of light? Otherwise, will just played the video of the experiment.
Both can go arbitrarily wrong.
The first question would be whether you can find suitable functions in Avisynth or a finished filter or persuade someone can take a writing (a great www.doom9.org Avisynth corner, the Forum has neuron2.net synonymous with smart minds ...) has .
The next question is whether such a calculation is linear at all to luminance of the monitor or the experimental design.
Clever, it would have been if the advice my grandmother had been applied: Child FONG no television in the dark. That is, if the light fluctuations were hidden by the video through a suitable backlight of the experimental room. Far below, in this case the issue still lurks transient adaptation (no, do not panic, I'm not so smart, I just read it at Wikipedia ;-)
Specifically considered, I would rather make a real light measurement in the experiment. For this you need a light meter with data interface to the PC and with a sampling rate below the response of the pupil (Pupillenreflex. .. How quickly anyway? (<1s, hmm) and of course time to play the video in real time or in Slowmo is if you find only slower exposure meters).
Good luck
The fat Stefan
Antwort von inwa:
for an experiment, I have measured the Pupillenvergroesserung with people while they watched a video.
The keyword meter and test set-up has been already mentioned. Therefore, I have only myself eigendlich questions:
What do you want eigendlich with the purpose to the experiment? Do you study psychology?
Antwort von colacao:
Hello,
tausend dank erstmal s.Stefan for the answer. Avisynth times I've watched and vllt. What would be your proposal synonymous for me, synonymous if I'm not sure if I can implement it ...
Using eye tracker measured the pupils, but that was made about 50 times per second. The tester sat approximately 1.5 m of a large television from ambient light and as we were "normal" daylight (if I may so describe the times). Using these measurements, we wanted to determine when and who arouses in the video was to close on his emotions, I (less the issue here, just anyway from simple, if this is not the synonymous). Now we wanted to measure the light from the film to subtract the values of the pupils, since pupils Change customer indeed synonymous depending on the light so that we examine "almost exclusively" the effect of emotion can. At least let's try and see if this goes ...
So I'd like to measure the light in the video, which have seen the testers, because what distinguishes the different sequences of the film. The ambient light has not changed so drastically. The idea with the light metering, we have taken only just launched, so I'm an absolute amateur.
Whether the scientific welfare as such is well thought out, is already an issue, but we want it a try if that can help us in the future.
and @ Inwa: I do not study psychology, but it is connected with. Uni brash s.der in video game development, and meaning and purpose is to find methods to measure emotions in players as simple as possible, in developing the game design.
If anyone has any criticism or suggestions, please bring it on them.
Gruss.
