Spectroradiometers are used to determine the spectral characteristics and brightness of radiation sources, which are widely used in many different fields. This study is based on the circular variable filter type spectroradiometer, where the wavelength transmitted by the main spectroscopic component, the circular variable filter is linearly related to the angle, and the spectroradiometer is constructed with a unit detector. This type of spectroradiometer has the advantages of a wide spectral range and a wide temperature range for the target, so it has a wider range of applications. However, there are fewer studies on circular variable filter spectroradiometers in China and abroad, and the development of domestic machines for circular variable filter spectroradiometers is gradually being carried out in China. Radiation calibration is the process of converting the original signal measured by the instrument into a physical quantity with practical significance. The main methods of radiation calibration for infrared spectroradiometers are currently the single point method, the two points method, and so on. The single point method is suitable for cases with low resolution and a small amount of spectral measurement data. The two-point method is suitable for situations where the instrument has good linearity, and the number of measurement points is high. Due to the wide operating band of the circular variable filter type infrared spectroradiometer and the wide range of the target temperature, which causes non-linearity problems, the traditional two-point calibration method cannot achieve accurate radiation calibration. In this paper, a divisional linearity-based responsivity radiometric calibration method is proposed to solve this problem.

The radiometric calibration of circular variable filter spectroradiometers is based on the divisional linearity method, which is used to solve the non-linearity problem of this type of spectroradiometer due to the large temperature range of the measurement target and the wide operating band. The main technical principle is to divide the temperature interval of the target to be measured into several subintervals, collect the measured spectrums corresponding to several different temperature blackbodies in the target temperature interval, and calculate the responsivity function at each temperature. During the infrared spectroscopy measurements, the target spectrum is compared with the spectrums of different temperature points recorded in the interval to determine the upper and lower limits of the temperature subinterval to which the target to be measured belongs. Based on the responsivity function calculated for the subinterval, a linear interpolation is performed to find the responsivity function of the target to be measured for radiometric calibration. In addition, external ambient temperature variations, atmospheric disturbances, and the instrument's thermal radiation are taken into account in the calibration process.

In this paper, we propose a divisional linearity-based responsivity radiometric calibration method, which can effectively solve the non-linearity problem caused by the wide wavelength range and wide temperature range of the target measurement by zoning the target temperature into sub-regions. We compare the difference between the measured calibration data and the theoretical Planck curve at different temperatures. Figure 12 shows the relative deviation of the radiometric calibrations of two detectors at different blackbody temperatures. Figure 12 shows that the relative deviations of the radiometric calibrations are better than 1% for most of the band intervals for both detectors. The large relative deviations in some bands are due to two reasons: 1) the low responses of the InSb detector in the 2.4-3 μm region and the MCT detector in the 13.5-14.3 μm region are due to the low signal-to-noise ratio of the collected signals in this region, which affects the calibration accuracy; 2) the InSb detector in the 4.2-4.5 μm region is due to the interference of CO₂ atmospheric absorption in this band. The interference of CO₂ atmospheric absorption exists. The experimental results show that this method can effectively meet the radiometric calibration requirements, and the calibration results are in good agreement with the theoretical values, with an equivalent temperature deviation of less than 2%.

The large temperature range and the wide operating band of the circular variable filter spectroradiometer make for a significant non-linear response in the radiometric calibration process. Different temperature targets also have different degrees of responsiveness, so the traditional two-point method does not work well for radiometric calibrations. In addition, external ambient temperature variations, atmospheric disturbances, and the instrument's thermal radiation are taken into account in the calibration process. In this paper, a divisional linearity-based radiometric calibration method is proposed, which can effectively solve the non-linearity problem caused by the wide wavelength range and temperature range of the target measurement by zoning the target temperature into sub-regions. The experimental results show that the method can effectively meet the radiometric calibration requirements, and the calibration results are in good agreement with the theoretical values, with an equivalent temperature deviation of less than 2%. The zoned linear radiometric calibration method in this paper is also applicable to other spectroradiometers of the spectral type to solve the non-linearity problem caused by the measurement of targets with a wide wavelength and temperature range.