Correction of HDR images

hdri
1

Hi list,
I wanna compare the error between the sensor-measured and HDR-derived vertical illuminance to check the validity of the HDR images. I finished the whole procedures by refering to this paper:
https://doi.org/10.1080/15502724.2019.1684319
However, there is something weird about “evalglare” with invaild exposure. This caused me unable to get the illuminance from the HDR images. I checked the original image and the corected images. It seems that “pcompos” command has changed the exposure. Any suggestion is appreciated!
Besides, I used this version of Radiance:

tmp233B
tmp233B735×326 60.5 KB

7w1.hdr
image
image919×616 54.4 KB

7w1_c.hdr

image
image669×669 63.3 KB

image
image734×185 26.7 KB

image
image704×150 18.4 KB

image
image728×401 45.9 KB

1 Like
2

I think evalglare objects to seeing an EXPOSURE= line in the previous input headers, even if there is a valid EXPOSURE= line in the final (left-justified) header portion. This is a problem for both pcompos and pcomb (even using -o option).

I think you can avoid the complaints by using the “-h” option to pcompos and pcomb, which leaves off information from the previous headers. You should try that.

Best,
-Greg