针对空间目标与地基望远镜红外成像传感器终端之间的不确知参量将降低双波段比色测温法求解精度，且精度影响程度未知，假设目标为灰体，对包含不确知参量的最大似然估计函数关于发射率求偏导，建立基于红外探测器测量电子数的双波段比色测温数学模型，并进行双波段比色测温法的蒙特卡洛仿真实验与精度分析。对于大气透过率的估算，提出应用红外自然星体的大气透过率现场标校方法。空间目标温度反演精度与成像探测器的信噪比、大气透过率、地球热辐射以及波段之间的发射率差异等未知参量的估算精度有关。结果表明：信噪比高于20，波段之间发射率差异小于0.03，地球热辐射预测精度优于50%，大气透过率预测精度优于10%时，比色测温法优于40 K的温度估计精度。

The unknown parameters between space target and ground-based telescopes infrared imaging terminal reduce the dual-band pyrometry accuracy, and the influence degree is unknown. Assuming that the target is gray body, the partial derivative of maximum likelihood estimation function about emissivity is solved. The dual-band pyrometry mathematical model based on infrared detectors to measure the electron number is established, and the Monte Carlo simulation and precision analysis of dual-band pyrometry is performed. The field calibration method of atmospheric transmittance prediction based on infrared natural star is proposed. The inversion precision of space target temperature has a close relationship with the imaging detector signal-to-noise (SNR) ratio, the earths thermal radiation estimation accuracy, atmospheric transmittance and emissivity difference between bands. When the SNR is above 20, emissivity difference is less than 0.03, the earth radiation prediction accuracy is better than 50%, and atmospheric transmittance prediction accuracy is better than 10%, the temperature estimation precision of dual-band pyrometry is better than 40 K.