Re: IR scenes

New Message Reply Date view Thread view Subject view Author view

Angus Dorbie (dorbie++at++sgi.com)
Wed, 28 Oct 1998 11:46:22 -0800

It's important to separate the computation of the physics of a
thermal simulation and the ability to render the result well.
As you say some aspects can be computed in the gl but I don't
see a graphics library as the place to tackle the other kinds
of heat transfer.

With the GL you have so many tools, including multipass
rendering of the same geometry with different state and direct
access to texture imagery, blend equations, TLUTS and texture
environments, component masks and color matrices that you have
more than ample tools to render thermal scenes well.

It is possible to build a database with emissive properties and
MultiGen has limited support for this.
It is also possible to do this for a specific sensor so that
the radiance of the database approximates the sensor result
given more complex spectral analysis, something like IRgen will
attempt this with an OpenFlight model and build it into the
OpenFlight fields. There are products available through
MultiGen|Paradigm and others which attempt to simulate the
thermal equations in real time according to events in the
simulation and then render that data correctly for a specific
sensor. In both cases the calculations require more information
about the material properties of a database.

Cheers,Angus.

Russ Moulton, Jr wrote:
>
> drb wrote:
> >
> > Dear performers,
> > How do we make a visual database IR compatible? Using a modelling tool
> > like MultiGen, is there a way to assign IR properties to polygon, which
> > will behaviour built in? Any one tried any other methods?
> > Thanks!!!
> > Balaji
>
> Balaji,
>
> Assuming you want an OpenGL-compatible, time-dynamic database,
> you are generally stuck with a set of properties (ambient,
> diffuse, specular, emissive) for modeling the reflected and
> self-emitted signature components in a particular spectral band
> (like visible, near, mid or longwave-IR).
>
> OpenGL (and thus Performer) handles dynamic reflected signature
> phenomenology very well, like visible-band light sources
> interacting with visible-band reflective surfaces for example. Indeed,
> the OpenGL lighting model (often implemented in hardware) computes
> the total reflected component "on-the-fly". You assign the appropriate
> position/in-band intensity light source, and assign the
> appropriate in-band reflective properties (ambient, diffuse, specular)
> on the vertices of your database and it does the rest.
>
> Unfortunately, you are "on your own" for the emissive component -- you
> have to assign that yourself.
>
> For the visible (e.g. human eyeball, TV cameras) and near-IR band
> (e.g. Night Vision Goggles), you can ignore the neglible emitted
> component
> for terrestrial and man-made scene objects (provided
> they don't get too hot, like bulb-filaments). You must of course,
> appropriately assign the time-dependent source intensities (sun, moon,
> skyshine, earthshine, etc) and reflectivities for the band you want.
>
> But in the mid-IR and longwave-IR, you must have a means to compute
> the dominant emissive component. This component is driven by the
> object's temperature and in-band emissivity. In general, to get a
> dynamic
> temperature, you need to solve the heat diffusion equation,
> which takes into account the thermophysical and surface optical
> properties of the object, as well as the incident irradiance.
> (thermal conductivity, specific heat, density, coefficient of
> convection, solar absorptivity, longwave emissivity, solar insolence,
> longwave irradiance, internal heat sources/sinks, and evaporative/
> condensating heat transfer).
>
> There are a myriad of methods and shortcuts to address the
> emitted component, depending on your application's real needs.
> One technique, for example, adapts the lighting model and unique
> OpenGL properties you would assign to produce dynamic (pre-computed)
> IR signatures.
>
> Hope this is helpful in understanding the issues involved.
>
> --
> Russ Moulton, Jr.
> JRM Enterprises, Inc.
> Voice:540-785-4585 Fax:540-785-4586 russ.jrm++at++eosoft.com
>
> =======================================================================
> List Archives, FAQ, FTP: http://www.sgi.com/Technology/Performer/
> Submissions: info-performer++at++sgi.com
> Admin. requests: info-performer-request++at++sgi.com

-- 
"Only the mediocre are always at their best." -- Jean Giraudoux 


For advanced 3D graphics Performer + OpenGL based examples and tutors:
http://www.dorbie.com/

New Message Reply Date view Thread view Subject view Author view

This archive was generated by hypermail 2.0b2 on Wed Oct 28 1998 - 11:46:31 PST

This message has been cleansed for anti-spam protection. Replace '++at++' in any mail addresses with the '@' symbol.