Hello Tobias,

Unfortuately you will need to crop the image using some of the Radiance tools.

First you have to set the image exposure to 1 using ra_xyze,
ra_xyze -r -o in.hdr > exposure.hdr

Then use pcompos to do the cropping,
pcompos -x 450 -y 450 exposure.hdr X Y > cropped.hdr

Where X and Y are the anchor coordinates from the lower left corner of the image, if I recall correctly. I always end up having to tweak them until I get it correct. Some examples can be found by searching the mailing list.

Finally, this process will eliminate the view information you have, so it will need to be put back into the header.

Best,
Alstan

On Tue, 29 Jul 2014 08:01:00 -0400, Tobias Porsch <[email protected]> wrote:




Hello again,






Thank you all for the many hints. I tried it out and it works now.



But anyhow I have a following problem now.






When I do insert the following lines to the header for the description of our used angular fish eye >lens, radiance tools like 'evalglare' (the author is J. Wienhold) does calculate wrong results.






#?RADIANCE



FORMAT=32-bit_rle_rgbe



VIEW= -vta -vv 180 -vh 180 -vp 0 0 0 -vd 0 0 1 -vu 0 1 0



EXPOSURE=1.79e+02






-Y 1040 +X 1392









I do assume this is due to the fact that the circular field of view for our fish eye is just a part >of the rectangular aspect ratio of the image.



For example the center of the circular view is x = 700px; y = 500px and the diameter = 450px.



Is it possible to tell this within the header?






Another possibility could be to crop the image according to this circular field of view. Is this >possible?






Kind regards



Tobias














Von: Gregory J. Ward [mailto:[email protected]]


Gesendet: Mittwoch, 23. Juli 2014 17:57

An: High Dynamic Range Imaging

Betreff: Re: [HDRI] Question about header information of HDRI









Well, you asked at the right place! The Radiance header format is simple, a set of newline ('\n') >separated comments, some of which are of the form:










VARIABLE=value










A space may appear after the equals ('=') but rarely/never appears before it. Other lines in the >header do not have this form, and may be anything. However, the first line identifying the file >should be:











#?RADIANCE










The "RADIANCE" portion above is traditional, whereas the "#?" sequence is the so-called "magic >number" for Radiance pictures. Below is a typical information header:










#?RADIANCE





CAMERA= NIKON CORPORATION NIKON D700 version Ver.1.03





Photosphere created HDR image from '_DSC7719.JPG' '_DSC7720.JPG' '_DSC7721.JPG' '_DSC7722.JPG' >'_DSC7723.JPG' '_DSC7724.JPG' '_DSC7725.JPG' '_DSC7726.JPG' '_DSC7727.JPG'





EXPOSURE=4.887013e-03





VIEW= -vtv -vh 51.765083 -vv 71.303574





OWNER= Greg Ward





CAPDATE= 2014:02:25 09:49:32





GMT= 2014:02:25 08:47:39





LATLONG= 46.948730 7.451972





FORMAT=32-bit_rle_rgbe











------





The header is terminated by an empty line (two newlines together "\n\n"). The following line after >the empty line is very important, it contains the image resolution. It is typically in the form:










-Y ysiz +X xsiz










where xsiz and ysiz are the integer number of pixels in the image. This ASCII string is terminated >by another newline ('\n'), immediately followed by the raster image data in the order indicated, Y >(vertical) major sort, with Y values decreasing as measured from the bottom-left of the image. This is explained in more detail in the >relevant section (pp. 28-35) of the Radiance file formats document, located here:










              http://radsite.lbl.gov/radiance/refer/filefmts.pdf










I see Lars has provided you with the same information. I hope this is enough to get you started. > Please note the importance of the EXPOSURE= line in the header and the 179 lumens/watt standard >conversion factor in creating a properly
calibrated picture.










Let me know if you have further questions.










Best,





-Greg













From:
Tobias Porsch <[email protected]>




Subject:
[HDRI] Question about header information of HDRI




Date:
July 23, 2014 8:27:01 AM PDT








Hi,






I'm writing you concerning my current project.



I'm working for TechnoTeam company. Manyfacturer of imaging luminance measuring devices (camera >>photometer).



Our aim is to export our own image file format (pixel-wise HDR float values) into the Radiance rgbe >>file format.



The aim is to give radiance users access to the measured luminance information combined with our >>information about lens configuration (geometrical data).



This may of interest for the "glare" community when analyzing the glare using radiance tools.





Anyway, writing the rgbe information is no problem.






But what about the header structure of the radiance HDRI format?



I'm a little confused about it and can't find a documentation structure how it is organized?



Unfortunately I'm no radiance user for carrying out own radiance projects to study it?







With my best regards



Dipl.-Ing.
Tobias Porsch






TechnoTeam Bildverarbeitung GmbH

Werner - von - Siemens - Strasse 5
98693 Ilmenau

----------------------------------------------------------------------------

Email: [email protected]

Phone: +49 3677 4624 0






TechnoTeam Bildverarbeitung GmbH



Geschäftsführer: apl. Prof. Dr.-Ing. habil. F. Schmidt



Standort: Ilmenau



Handelsregistereintrag: Jena HRB 300912












_______________________________________________
HDRI mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/hdri

Reply via email to