Shutter-less infrared cameras based on microbolometer focal plane arrays (FPAs) are the most widely used cameras in thermography, in particular in the fields of handheld devices and small distributed sensors. For acceptable measurement uncertainty values the disturbing influences of changing thermal ambient conditions have to be treated corresponding to temperature measurements of the thermal conditions inside the camera. We propose a compensation approach based on calibration measurements where changing external conditions are simulated and all correction parameters are determined. This allows to process the raw infrared data and to consider all disturbing influences. The effects on the pixel responsivity and offset voltage are considered separately. The responsivity correction requires two different, alternating radiation sources. This paper presents the details of the compensation procedure and discusses relevant aspects to gain low temperature measurement uncertainty.
This paper concerns with the problem of disturbing radiation derived from the interior of radiometric
microbolometer-based infrared cameras. The amount of internal radiation depends particularly on the ambient
temperature. Variation of ambient temperature leads to a change of the temperature distribution inside the camera. The
approach proposed here is determining the disturbing radiation without using a shutter by measuring the internal thermal
state with several temperature probes and deducing the disturbing radiation flux. Because of this discrete temperature
measurement it is not possible to determine the present thermal state of the camera interior as precise as performing a
shutter process. Therefore, the position of the temperature measurement is crucial for the significance of the relation
between measured temperature and disturbing radiation flux. Furthermore, the transient thermal behavior during a
cooling or heating period of the camera enclosure is a non-ergodic process . Two approaches facing these problems
The first approach is based on the usage of more than one temperature probe at different positions inside the camera.
Each position of temperature measurement has its own characteristic of heat conductance and convection parameters.
Therefore, the low-pass behavior and the corresponding response time of the measured temperature in relation to the
ambient temperature differ. Developing a thermal model using different probes with a higher significance of the transient
thermal trend reduces the calculation uncertainty.
A second approach is to separate the transient and the steady-state behavior of the calculation model. If the camera is
able to follow a slow change of ambient temperature completely, then it stays always in steady state and the process is
ergodic. Only in case of an abrupt change of ambient temperature the thermal behavior leaves the steady state and a
transient correction factor is necessary. This factor has to take the history of the measured temperature into account.