Aerosols and clouds in the atmospheric environment are important factors affecting the image quality from high-altitude platforms. Aerosols exhibit relatively strong polarization effects compared to ground reference objects, especially under adverse conditions such as sand and dust storms. When aerosols are mostly concentrated in low altitude areas, it has a significant impact on the image quality of aviation platforms. To adapt the application requirements of atmospheric correction for aviation platform images, multiple spectral bands of polarization radiation information in the atmosphere are obtained from the visible to shortwave infrared bands, which can make to achieve spatial and temporal change of aerosol and water vapor detection, and retrieve atmospheric aerosol and water vapor characteristic parameters to solve the difficulties of atmospheric correction.The Aviation Polarization Radiometer (APR), developed by the Anhui Institute of Optics and Fine Mechanics (AIOFM), Chinese Academic of Sciences (CAS), has the characteristics of compactness and lightweight for aerial applications, which is mounted on an aviation flight platform. The Aviation Polarization Radiometer is designed to detect aerosols, land surfaces, and sea surfaces under aerial flight modes, which need to have a certain dynamic range to adapt and obtain effective data for different targets.Firstly, a systematic description is provided for the instrument. The airborne polarization radiometer adopts a design scheme of a channel type optical system combined with a unit detector. The APR achieves wide spectral coverage and high-precision information acquisition. The instrument utilizes the filters to achieve band division, there are four polarization spectral channels from visible to near infrared, namely 490 nm, 670 nm, 870 nm, and 1 610 nm. Each polarization spectral channel is detected through a combination of three filters and three polarizers for spectral and polarization information detection. Secondly, to ensure the availability of instrument data and the accuracy of parameter inversion, the laboratory calibration is conducted after the instrument is assembled and adjusted. The absolute radiometric calibration and polarization measurement accuracy are verified for the APR.In order to ensure the accuracy of the absolute radiometric calibration, the calibration experiment is performed using a radiometric calibration system based on a spectrometer. This calibration system comprehensively considers the uncertainty in absolute response, which meets the requirements for radiometric calibration accuracy. In addition, an adjustable polarized light source is utilized to verify whether the performance of the instrument fulfills the requests in various polarized spectral bands. This validation ensures the accuracy and stability of the instrument's measurements in different polarization bands, meeting the expected performance criteria. At the same time, the instrument's response suffering to out-of-band effects in the 490 nm wavelength is analyzed and validated.The measurement results indicate that the absolute radiometric calibration uncertainty of the APR is less than 2.51%. When the polarization degree of the input polarized light is 20%, the polarization measurement accuracy in various polarized spectral bands is lower than 0.16%. At the same time, an analysis is conducted on the unsatisfactory polarization accuracy in the 490 nm wavelength band, and a monochromator system is used to verify its influence on out of band response. Through comparative experiments, it is confirmed that by removing the out-of-band response influence on the polarization channel at 490 nm band, the polarization accuracy is improved from 2.29% to 0.06%. This improvement provides support for subsequent product optimization, such as optimizing the filters' cut-off wavelength to enhance the instrument performance.