I’m hoping to learn if Radiance is the right tool for this research project and hopefully get some guidance from the community on how to get started if it is.
There is a type of camera that uses long wave infrared, which is further away from the visible spectrum than near-infrared, which has already been a topic here. The sensor plates of these cameras are not photon receptors, but thermal resistors because the wavelength is the same as thermal radiation.
The first objective is to do a sun-lit render of an object, using the long wave infrared wavelength.
The second objective is to make parts of the object glow as though they are hot and render both.
The third objective is to somehow simulate things like lens-flares, glare, and bloom.
For this experiment, I’m keeping some things simple by assuming the object being rendered is a metal box floating in the vacuum of space, lit only by the sun and heaters built on it.
What’s important is that the results could be verified by real-world experimentation, which is why Radiance was my first choice for this project.
I have some vague ideas on how this could be done, but perhaps it’s best not to steer the conversation towards a particular direction right away.
Thank you all! Would love to get your feedback!
Your topic seems vaguely related to this thread. I don’t suppose you are working on different aspects of the same problem, are you?
In any case, Radiance is not really designed for long IR simulation inasmuch as materials are either emitters or passive reflectors/transmitters of radiation, not both. You can mix the “glow” type with other types, such as “plastic” to get some combined behavior, but it is tricky to do so correctly. Also, Radiance has no internal mechanisms for tracking temperature, thermal mass, conduction, convection, radiative losses and gains, or even black body spectra. All of that would be your job as well.
Assuming you are up for that, it is possible to explicitly model longwave radiation using the latest spectral rendering additions in the HEAD release found here. You would need to specify a range that included visible light from at least 468nm, and the photon-mapping add-on is not yet functional in this prerelease, but it is possible.
Regarding lens effects such as flare and bloom, we don’t have any built-in simulation models for these effects, and I am also unsure if or how thermal wavelengths differ from visible.
Given all that, you are welcome to give it a try, but there may be better-suited software out there. It’s not really my area of expertise; I just didn’t want your query to go unaswered.
Cheers,
-Greg
Thank you for the swift reply!
I will look into the glow type and the spectral rendering additions you mentioned.
If we were able to generate a grey-scale temperature map with another program and apply it as a texture, do you think there would be a way to make it glow according to the bright and dark areas of the texture?
You would need a tool that converts the thermal map into an HSR (hyperspectral radiance) image with the appropriate thermal range and black body emission spectra, then apply this to a surface modified by a glow or a mixfunc with a glow and some reflecting material, with the glow then modified by a specpict pattern. It is possible, but no one has done it, before.
Cheers,
-Greg