Sources1

The scene consists of an array of point sources illuminating a simple, flat surface. The scene has two rows containing 3 types of sources:

The first row is configured using the GDB file mechanism and the second row is configured using the GLIST mechanism.

The following demos, manuals and tutorials can provide additional information about the topics at the focus of this demo:

A detailed description of how sources are configured is discussed in detail in the User-Defined Sources Manual. In general, a user-defined point source is a special geometric entity defined by a point in space with a pointing direction (in case the source has directional shaping configured) with the radiometric properties defined via a material with special "source" properties.

This section highlights key files important to the simulation.

The output radiance image is displayed below. The top row of sources were configured via the GDB file mechanism and the bottom row via the GLIST <basesource> mechanism. The 3 different source materials are arranged horizontally with the 300W 3,000K (left), 50W HPS (middle) and 250W Hg Vapor (right). Note that the spot size of the left source (300W, 3000K) appears smaller because it’s brightness and efficacy are lower than the other two types. The color difference is a direct result of the spectral power distribution described by the respective Radiant Intensity files.

Since this scenario is simple, it allows us to quickly check the DIRSIG numerical calculation against an external analytical solution. For this verification, there is a separate set of simulation files that look at the radiance reflected off the background directly below the 300W 3000K source. The source shape coefficient is 0, which makes this an omni-directional point source. Therefore, the irradiance (E) arriving at a surface r meters away from a source with radiant intensity I follows the Inverse Square Law:

\begin{equation} E = \frac{I}{r^2} \left [ \mathrm{\frac{W}{m^2}} \right ] \end{equation}

The illuminated surface is a Lambertian reflector, therefore the surface leaving radiance (L) can be computed as:

\begin{equation} L = E \left ( \frac{\rho}{\pi} \right ) = \frac{I}{r^2} \cdot \frac{\rho}{\pi} \left [ \mathrm{\frac{W}{{m^2 sr \mu m}}} \right ] \end{equation}

At 0.6 microns, the radiant intensity is 8.456 (see 300w_3000k.int) and the source is 2 meters above the 18% Lambertian reflector surface:

\begin{eqnarray} L &=& \frac{8.456 \mathrm{\frac{W}{sr \mu m}}}{(2 \mathrm{m})^2} \cdot \left ( \frac{0.18}{\pi} \cdot \frac{1}{\mathrm{sr}} \right ) \nonumber \\ &=& 1.2112 \times 10^{-1} \left [ \mathrm{\frac{W}{{m^2 sr \mu m}}} \right ] \nonumber \\ &=& 1.2112 \times 10^{-5} \left [ \mathrm{\frac{W}{{cm^2 sr \mu m}}} \right ] \nonumber \end{eqnarray}

Since the atmosphere in this simulation has no extinction or path radiance, the sensor reaching radiance is the same as the surface leaving radiance. The simulation that attempts to isolate this source can be run using the test.sim scenario. It uses a small 33x33 camera with raw spectral radiance output so that we can easily extract the reflected radiance directly under the source.

The peak radiance should be the point on the surface closest to the source. Due to our viewing geometry, this will be the center pixel of the output image. This value can be found by loading up the 0.6 micron band (Band #21) in the graphical image viewer and looking at the reported Data Value at the center of the image. Or the image_tool utility can be used via the command-line to verify the maximum value is a match:

The following simple variations of the simulation can be performed: