![]() |
Re: camera-ui2 (now a part of CSSU) (updated 09. May)
Quote:
To quote "Predator": "You're one, ugly... mother****er." (Your N900, of course.) |
Re: camera-ui2 (now a part of CSSU) (updated 09. May)
Maybe you can "cut" or remove the protective "glass" on N900 camera and replace it with common glass... http://en.wikipedia.org/wiki/Cover_slip
It may turn your camera unusable... or it may enable it to take Night Vision Shots... :P |
Re: camera-ui2 (now a part of CSSU) (updated 09. May)
Thanks, but it doesn't matter. IR filtering is done (in our case) directly on sensor, probably hardware controlled.
/Estel |
Re: camera-ui2 (now a part of CSSU) (updated 09. May)
Ay? It will be an optical lowpass filter surely...(i.e. A piece of glass/plastic)
|
Re: camera-ui2 (now a part of CSSU) (updated 09. May)
Hm, in such case, need to check it again. I must admit - I based assumption about hardware'ish filtering on discussion from fCam API forum (can't find link now), where someone gave up due to inability of enabling IR full-pass. That's why I focused on building multi-IR-diode setup powered from N900 battery, and later, from USB port.
The known "joy trick" is watching IR diodes light from remote controllers on N900 viewfinder - so, our filtering isn't very restrictive, yet, too high to allow true short range nightvision. Does anyone have idea, which part exactly can be IR filtering? Could it be plastic part that we see from outside, or something between lenses and sensor, like strange1712 suggested? AFAIK (again, based on talk with some photo guy, not my own research) removing IR filtering would case problems with focus, when taking regular photos. So, *if* filter is really accessible, we would need to setup some "switching" mod. Quite complicated, but it is nothing compared to joy of having short-range high resolution Night Vision option in our lovely device, almost for free (bunch of cheap IR diodes, and some modding work). /Estel |
Re: camera-ui2 (now a part of CSSU) (updated 09. May)
Correct, for NIR photography in the old days lenses would needed to be focused differently, using the famous red dot mark,
http://www.maxmax.com/ir_techniques.htm http://www.luminous-landscape.com/re...d%20dslr.shtml ("More on Infrared Focusing") http://photo.net/black-and-white-pho...g-forum/00OIsi That's why i never got to buy a newer camera after my PowerShot G1 - it was too simple to shoot NIR, even with an B+W 093 filter. |
Re: camera-ui2 (now a part of CSSU) (updated 09. May)
Quote:
|
Re: camera-ui2 (now a part of CSSU) (updated 09. May)
I don't see any reason, why it should be hardware restricted. yet, for sure You can't achieve it with camera-ui - it's using what libraries provide, so no fine-pitch exposure control.
With fCam libraries, it should be 100% possible (fixme?). /Estel |
Re: camera-ui2 (now a part of CSSU) (updated 09. May)
I'm lost.
Estel, what exactly are you trying to achieve with your N900? I understand the tripod part but the rest... Please excuse my ignorance. |
Re: camera-ui2 (now a part of CSSU) (updated 09. May)
Few independent things, lest say, few projects.
One, that I haven't considered seriously until recently, is short-range Night Vision. Google for original idea - when it comes to our device, the point is that our sensor is perfectly capable of catching full InfraRed range, yet it is limited by hardware. Until recently, I though it is hardcoded in sensor controller (i.e kinda software'ish limit, but impossible to bypass without hardware change = just lik e hardware limit). Ho ever, strange and other suggested, that IR is filtered rather by small plastic part - hardware filter. Just as proof of concept, some time ago I prepared set of IR diodes, that may be powered from both USB (via hostmode) and battery itself (using dr-frost-dk's modded battery with external plug). The power of InfraRed light is enough to light up 15-30 meters in full darkness (shorter when using USB hostmode, longer, due to higher power, when using N900 battery itself) - of course IR light is invisible for human eyes, but is perfectly visible for our sensor, if we're able to remove IR filtering plate. Now, why the plate is there at all? When not filtering IR, light spectrum on photo is different, and algorithms used by autofocus are not working (in our case - as don-falcone pointed out, there are workarounds for good'old standalone cameras). Generally, some photographers remove IR filtering to get some very interesting looking photos even in daytime - after all, we also have manual focus - but, my main goal is to enable Night Vision mode, where we can see things in the dark, without actually lighting surrounding up by light visible to human eye. Particularly, You look at screen, and You see things, that are not visible without screen. I don't know how to write it more clear - You just need to have basic idea what NightVision device is. This can be (probably) achieved quite easy right now, with one drawback - we would lose ability to take normal daylight photos (due to broken autofocus and bizarre colors). This doesn't match my 'aesthetic' ;) so I'm designing hardware mod to physically swap IR-filtering plate, and 'dummy' regular glass/plastic plate (without filtering capabilities). Of course, I first need to check if the outer plate (that is accessible) is really *this* plate that perform filtering. --- Now, second thing, that tecs mentioned in his post, is exposure time (not related to IR and NightVision). Max shutter open time - achievable by fCam libraries - is 1 second. It's very long for regular photos, but using tripod, and taking night shots (normal, not infrared), it is many times - if not most - desirable to have longer shutter open time. This way, You can make very good quality photos of static objects, even in *very* low light enviroments, or, interesting photos of moving objects (like parallax of stars). /Estel |
All times are GMT. The time now is 05:45. |
vBulletin® Version 3.8.8