Test in Japan (2007)
Taking Images and Enhancement of Dynamic Range
- An image taken by the camera consists of 8-bit data.
- The sensitive brightness range of the camera can be changed by setting the parameter ``BR''.
- Since the state of the whole sky/clouds does not significantly change in timescale of < 30 seconds, the effective dynamic range can be enhanced (to ~9.5-bit) by taking images successively with several different settings of the camera and then combining them (see below).
Test Run at the Ibaraki Univ.
A student (W. Kimura) is setting up the camera on the rooftop of our university building
From another angle
Example of the images
- (1)-(7) in the below figure are the images with different settings of the paramete "BR". (see the figure caption below for the more detail). Each of them consists of 8-bit data. Note that bright area in the bottom-middle in each panel corresponds to the emission from the temperature standard.
- After applying some appropriate offset to each image, all of them combined into a single "averaged" image (the panel "Ave"). After masking the non-sky area, the value is converted into an emissivity map by using the value of the temperature standard, which is assumed to be blackbody (the panel "in \epsilon").
Detailed explanation of the above figures
Raw data for every single time consist of 7 images with a different brightness parameter of the infrared camera ("BR" = -10,0,4,8,12,16,20). The parameter "BR" gets larger, the camera gets more sensitive to the difference in less bright regions. We used the image of "BR"=-10 (1) to determine the brightness level of the temperature standard, which is usually much brighter than the sky and clouds. Other six images (BR= 0 -- 20; (2)-(7)) are for the sky.
- Zipped Fits file of the above data is available.
- GIF animation; you can find clouds as well as the Sun...
Calibration Curve
- We determine the calibration curve by taking images of stuffs with various temperatures (-10 to 60 deg Celsius); we used a high-temperature cavity blackbody for a higher temperature and a cold metallic cup with salty ice for a lower temperature.
- We found that the output after applying some appropriate offset to unify number count with different camera setting is quite propotional to the Planck function integrated over 8-12 micron, as shown below.
- We exploit the formula below when we derive the emissivity of the sky.
