In: Wagner W., Szflcely, B. (eds.): ISPRS TC VII Symposium - 100 Years ISPRS, Vienna, Austria, July 5-7, 2010, IAPRS, Vol. XXXVIII, Part 7B 
60 
0.03 
0.02 
♦ ♦ 
co 
< 
0.01 ♦ 
♦ 
R 2 = 0.9855 
0 
0.07 0.12 0.17 0.22 
rms roughness, m 
Figure 6. Simulated and observed effects of cavities on 
thermal IR spectra. See text for details. 
Figure 7 shows another case where a form was created using 
plaster of Paris (CaS0 4 ’2H 2 0 ± CaC0 3 ) to facilitate model 
validation. The plaster has a single reststrahlen band centred 
near 8.6 pm. The form has two flat surfaces with different 
texture (roughness) shifted 7 cm and with a wall between them. 
The upper surface has hole that is 3 cm in diameter and 8 cm 
deep. In Figure 7, part a is a photograph of the Telops Inc. 
Hyper-Cam and the target; part b shows a hyperspectral image 
of the hole (and a crack that formed during transport); and part 
c is a Hyper-Cam image of the wall, and the upper and lower 
planes. Part d shows spectra taken from a Hyper-Cam image, 
and Part e shows the simulations at four spectral locations. The 
difference between the planes is the effect of roughness with the 
lower, rougher plane having a somewhat higher spectrum and a 
e) 0.96 
d) 0.96 
0.94 
¡6* 0.92 
LU 0.88 
0.86 
0.84 
-hole 
. upper plane 
• wall 
-lower plane 
8.5 9 9.5 10 10.5 
Wavelength, (tm 
9 9.5 10 10.5 
Wavelength, jjm 
Figure 7. Simulated and observed spectra for a constructed 
plaster-of-Paris form. See text for details. 
shallower spectral feature. The simulation did a good job on the 
spectral feature depth change, but not as good for the offset. 
The hole, both in the measurement and the simulation, had the 
highest emissivity spectrum, and very little of the spectral 
feature is present. The wall had a spectrum lower than the hole 
and with a clear but reduced spectral feature depth. The broad 
effects of radiosity were generally well simulated, impact of 
radiosity being proportional to view factors.