The blackbody is an ideal calibration target source for correcting the photoelectric conversion characteristics of infrared detectors. According to the theoretical radiation flux provided by the blackbody target surface, the detector can correct the quantitative relationship between the gray value of the output temperature field and the radiation flux, thus obtaining an accurate response gain coefficient. In order to meet the demand for real-time and accurate detection of targets with large temperature variations in fast-changing environments such as airborne, missile-borne, and space-borne scenarios, the wide-temperature-range blackbodies are typically vacuum-packaged by infrared windows to solve the problems of non-uniformity and low temperature dew and frost caused by large temperature differences in the environment. However, during calibration, the blackbody is affected by the attenuated target surface radiation and stray radiation from the window, there will be an error between the apparent temperature corresponding to the actual apparent radiation flux of the blackbody and the calibration temperature given by the measurement of blackbody surface. If the infrared detector still gives the calibration temperature according to the measured target surface temperature, it will affect the ideality of the blackbody and the calibration accuracy of the infrared detector. Most researchers consider the infrared window as a graybody and analyze the impact of environmental stray radiation reflected by the window on the infrared detector under high-temperature conditions, without systematically considering the influence of the spectral transmissivity, emissivity and reflectivity of the window on blackbody calibration technology. Therefore, in this paper, starting from the window spectral characteristics, aiming at the problem that wide-temperature blackbody packaged with infrared window affects the blackbody ideality, the influence law and internal causes of Ge and BaF₂ typical windows on the error between the calibration temperature and the apparent temperature corresponding to the actual radiation flux were analyzed under different environment temperatures and calibration temperatures. Furthermore, a calibration temperature correction method, which is associated with both the calibration temperature and the environment temperature, was proposed and experimentally validated.Firstly, a thermal-optical coupling simulation model for the extended source blackbody surface was established. By using Monte-Carlo ray tracing method, the diffuse gray blackbody surface, the infrared window surface with selective radiation and the environmental radiation parallel to the window were dispersed into energy-carrying sub-beams. The actual apparent radiation flux, apparent temperature, and the variation trend of the error between the apparent temperature and the calibration temperature of the blackbody were calculated by the integrating sphere under the conditions where the atmospheric transmittance band was 7.5 μm~13 μm, and the environment temperature and calibration temperature ranged from -50 ℃ to +50 ℃. The actual apparent radiation flux included the attenuated radiation flux of the blackbody target surface under the influence of window spectral transmission, as well as the stray radiation flux from the window's spontaneous emission and reflective environment. The results show that when the calibration temperature is between -50 ℃ and +50 ℃, if the calibration temperature is lower than the environment temperature, the apparent temperature is higher than the calibration temperature, and the positive error between the calibration temperature and the apparent temperature corresponding to the actual radiative flux increases with the increase of the environment temperature. Conversely, if the calibration temperature is higher than the environment temperature, the apparent temperature is lower than the calibration temperature, and the negative error increases with the decrease of the environment temperature. The maximum error for the blackbody packaged with a Ge window can reach -4 ℃ to +13 ℃, while for the blackbody packaged with a BaF₂ window, the maximum error can reach -6 ℃ to +20 ℃. This is because there the window's enhanced stray radiation and attenuated radiation from the target surface have a positive-negative game effect. Additionally, the average transmissivity of the BaF₂ window is lower than that of the Ge window, resulting in increased attenuation of the blackbody target surface and the stray radiation from the window.Secondly, based on the fitting of the error variation curve between the apparent temperature and the calibration temperature, the relationship between the error correction amount of typical Ge and BaF₂ window-packaged blackbodies and the calibration temperature change under different environment temperatures was obtained. In addition, the calibration temperature and apparent temperature measurement experiments of windowless open blackbody and Ge window-encapsulated blackbody were designed. Comparing the experimental value of the apparent temperature with the corrected calibration temperature, it was found that the effect of the blackbody calibration temperature correction was significant. For example, at an environment temperature of -10 ℃, the maximum absolute error between the corrected calibration temperature and the measured apparent temperature was 2.6 ℃, while at an environment temperature of 20 ℃, the maximum absolute error between the two was only 1.4 ℃.In this paper, through theoretical analysis, simulation correction and experimental measurement, it was found that the window has a positive-negative game effect on enhancing stray radiation and attenuating target surface emission radiation. Consequently, the larger the difference between the environment temperature and the calibration temperature, the larger the error between the apparent temperature and the calibration temperature. Additionally, the error increases as the average transmissivity of the window decreases. Furthermore, the effectiveness of the calibration temperature correction method was validated. The result shows that the maximum error between the corrected calibration temperature and the measured apparent temperature is reduced by more than 50%. This method provides a new solution for correcting the non-ideal characteristics of blackbody radiation sources with infrared window packaging and offers more accurate engineering guidance for the high-precision calibration of low-temperature infrared detectors.