[Radiance-general] DGP on 172 fish-eye image

Fri Mar 17 01:05:17 PDT 2017

Hi Jan,

Thank you for the clarifications. It all makes much sense and it¹s good to
know that the filling option was developed for those particular projection
conditions. I totally understand that if I¹m the only user with this problem
it does probably not worth your time to add that to Evalglare.
But that also means that there is no reason why I could not use the DGP
formula to make the calculation of a fish-eye image with these
characteristics elsewhere, correct?

I wanted to check with TechnoTeam, again, before getting back to you about
the specifications of my system. And indeed according to them my projection
is of an equidistant-type with a view angle of 172.

I noticed that my vertical eye illuminance measurements have a bit of a time
offset from the pictures, so really not appropriate to check for the
deviation. I¹ll get new values to test next week. But this also involves
using a newly created TechnoTeam¹s Radiance exporter which it is still under
validationŠso there are few things to get right in order to make this test
possible. I will get back to the list when I have a conclusion.

I¹ll see if I can still come to VeluxŠI had to make a choice because there
are so many other important events that one can attend this year. But it
would be really important to discuss this with you and the other PhDs.

Best Regards

Raquel

Raquel Viula

PhD candidate

TU Delft | Faculty of Architecture and the Built Environment |

Architectural Engineering and Technology

Julianalaan 134, 2628 BL Delft, The Netherlands | P.O. Box 5043 2600 GA
Delft

From:  Jan Wienold <jan.wienold at epfl.ch>
Reply-To:  Radiance general discussion
<radiance-general at radiance-online.org>
Date:  Wednesday 15 March 2017 19:40
To:  Radiance general discussion <radiance-general at radiance-online.org>
Subject:  Re: [Radiance-general] DGP on 172 fish-eye image

Hi Raquel,

 I agree that the outer 8° contribute only a little to the vertical
illuminance. The solid angle of that section is around 0.44sr (which is
around 7% of the total) and the cosinus for this flat angle of incidence is
also very low (in average around 0.07). so the overall impact of the outer 8
degree is less than 0.5%, assuming the luminances in that area are
comparable to the average of the image .

 Regarding the "filling" option in evalglare: this was implemented, because
there was a group using a fish-eye lens, where the projected circle was
larger in y-direction than the CCD-array. This function "fills up" the
missing parts with the last known value vertically. You cannot use it in my
opinion to fill up missing "ring" areas (to fill up between 172° to 180°).
Actually I could implement such a feature, but due to the low impact I don't
think it is necessary and so far no other users were missing such a feature.

 Are you really sure you have a equidistant-projecting lens with 172°?  The
equidistant types are  much less frequent on the market - most lenses cover
180° or more. 
 If you have a equi-solidangle fish-eye lens, you should correct the
projection as well.

 What you should definitely check is how close you get between measured
illuminance with the sensor and the calculated value from evalglare. If you
have a deviation there, you have a problem in the calibration (because the
influence of the outer 8% is really negligible, so you should nearly match
the value ).

 BTW: Are you coming to the academic forum of Velux May2nd? As far as I
know, there are at least three other PhD students there dealing with
HDR-cameras, daylight and glare and as far as I heard there are still free
spots.  I will be there as well.

 best
 Jan

 On 03/15/2017 10:31 AM, Raquel Viula wrote:

>   
>  
> Hi Alstan,
>  
> That¹s great to know, thank you. Yes, I have vertical eye illuminance
> measurements. I think it¹s probably a good idea to estimate what the error
> might be via simulation.
>  
> 
>  
>  
>  
> 
> Best Regards
>  
> 
> Raquel
>  
> 
> 
>  
>  
> 
> Raquel Viula
>  
> 
> PhD candidate
>  
> 
> 
>  
>  
> 
> TU Delft | Faculty of Architecture and the Built Environment |
>  
> 
> Architectural Engineering and Technology
>  
> 
> Julianalaan 134, 2628 BL Delft, The Netherlands | P.O. Box 5043 2600 GA Delft
>  
>  
>  
>  
> 
>  
>   
> From:  "J. Alstan Jakubiec" <alstan at jakubiec.net>
>  Reply-To:  Radiance general discussion <radiance-general at radiance-online.org>
>  Date:  Wednesday 15 March 2017 06:18
>  To:  <radiance-general at radiance-online.org>
>  Subject:  Re: [Radiance-general] DGP on 172 fish-eye image
>  
>  
> 
>  
>  
>  
>  
> 
> Hi Raquel, 
>  
>  
> 
> I would say yes. Just make sure that the view in the HDR header is appropriate
> or you specify it inline: evalglare -vta -vv 172 -vh 172 image.hdr
>  
> 
> This may slightly underestimate vertical eye illuminance calculated by the
> software, skewing your results a bit; however, 172 is close enough to 180 in
> terms of view angle that the error is likely minimal. Still, if you have
> vertical eye illuminance measurements, supply them inline with the -i flag as
> well.
>  
>  
> 
> Best,
>  
> 
> Alstan
>  
>  
>  
> On 3/14/2017 9:40 PM, Raquel Viula wrote:
>  
>  
>>  
>>  
>> Dear all,
>>  
>> I have a question regarding Evalglare. Hopefully this is the right mailing
>> list to ask. 
>>  
>> Can DGP be calculated for a -vta fish-eye photo capture that has a view angle
>> of 172 degrees? The manual says that Evalglare can be run for an image cut
>> horizontally given an externally measured vertical eye illuminance and Y max
>> and Y min coordinates but does not mention the possibility of using an image
>> that is also cut vertically, or shall I say, all around.
>>  
>> Thank you for your help.
>>  
>>  
>> 
>>  
>>  
>>  
>> 
>> Best Regards
>>  
>> 
>> Raquel
>>  
>> 
>> 
>>  
>>  
>> 
>> Raquel Viula
>>  
>> 
>> PhD candidate
>>  
>> 
>> 
>>  
>>  
>> 
>> TU Delft | Faculty of Architecture and the Built Environment |
>>  
>> 
>> Architectural Engineering and Technology
>>  
>> 
>> Julianalaan 134, 2628 BL Delft, The Netherlands | P.O. Box 5043 2600 GA Delft
>>  
>>  
>>  
>>   
>>  
>> _______________________________________________
>> Radiance-general mailing list
>> Radiance-general at radiance-online.orghttp://www.radiance-online.org/mailman/li
>> stinfo/radiance-general
>>  
>  
>  
>  
>  _______________________________________________ Radiance-general mailing list
> Radiance-general at radiance-online.org
> http://www.radiance-online.org/mailman/listinfo/radiance-general
>   
>  
> _______________________________________________
> Radiance-general mailing list
> Radiance-general at radiance-online.orghttp://www.radiance-online.org/mailman/lis
> tinfo/radiance-general
>  

-- 
Dr.-Ing.  Jan Wienold
Ecole Polytechnique Fédérale de Lausanne (EPFL)
EPFL ENAC IA LIPID

http://people.epfl.ch/jan.wienold
LE 1 111 (Office)
Phone    +41 21 69 30849

_______________________________________________ Radiance-general mailing
list Radiance-general at radiance-online.org
http://www.radiance-online.org/mailman/listinfo/radiance-general

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.radiance-online.org/pipermail/radiance-general/attachments/20170317/258c9c5d/attachment.html>