# Thermal and Multispectral Imaging

### a

Two solid spheres are in thermal equilibrium. They are surrounded by vacuum and there are no other radiators present. Compute T2, the temperature of the Sphere 2 given the data below. State any approximations made.

• Temperature: T1 = 5778 K.
• Radii: r1 = 695508000 m, r2 = 6357000 m.
• Emissivities: e1 = 1.0, e2 = 0.63.
• Transmittances: tau1 = tau2 = 0.
• Distance between the spheres: 149597870700 m.

Useful Matlab code: blackbody.m

### b

Sphere 2 is covered by a layer that reflects all radiation with wavelengths above 8 micrometers. Compute the new T2.

## Exercise 2: Multi/hyperspectral acquisition

Select one application, such as:

• Oceanography (sensor on satellites).
• Land-use classification (sensor on satellite).
• Land-use classification (airborne sensor).
• Crop monitoring (airborne sensor).
• Road/vehicle/pedestrian classification (car-mounted forward-looking sensor).
• Industrial inspection (camera mounted over conveyor-belt).
• Spectrography (one/few-pixel spectral imaging).
• ...

Describe why at least four of the six treated multi/hyperspectral acquisition techniques would be suitable/unsuitable for the selected applications (that is, advantages and disadvantages given the selected application).

## Exercise 3: Hyperspectral target detection

### a

Download the test data and take a look at the two hyperspectral images. What is the difference between the two, and which one is computed how from the other one?

Hint when visualizing: Bands 15, 7, and 3 corresponds approximately to red, green, and blue.

### b

In the folder "targets" there are spectra of a few objects avaialble. These objects are also present in the given hyperspectral images. Implement a hyperspectral target detector and try to find one (or more) of the targets in the one (which?) of the images. You can use one of the detectors from Sec 6.4 in the compendium, or from elsewhere in the literature.