为消除测距光谱通道位置和数量选取对氧气吸收衰减被动测距技术测距精度的影响,基于氧气吸收衰减被动测距技术的基本原理,分析氧气A、B吸收带光谱谱线特性。对于吸收带带肩上的光谱通道,利用蒙特卡罗法,以拟合非吸收基线与理想基线的误差平方和与相关度为指标,分析光谱通道位置和数量及拟合多项式级次对非吸收基线拟合精度的影响。对于吸收带带内光谱通道,分析不同光谱通道处吸收率大小对测距距离和测距精度的影响。结果表明:在综合考虑系统实时性和测距精度要求的情况下,A吸收带两带肩上光谱通道各为1个,B吸收带单带肩上光谱通道为2个,位置均宜选择在各带肩靠近吸收带一端;吸收带内光谱通道可根据测距任务中对测程和测距精度的要求灵活选择其数量和位置。因此,在无法一次性获取测距波段完整光谱曲线的情况下,单目多光谱被动测距系统采用较少的光谱通道和最简单的直线拟合方法,不仅可以保证系统的测距精度,而且能够减少滤波片更替和软件计算的时间周期,进而增强系统数据采集和解算的实时性。

In order to eliminate the influence of the location and number of spectral channels on the ranging accuracy of the oxygen absorption attenuation passive ranging technology, we discuss the basic principle of passive ranging technology based on oxygen absorption attenuation, and analyze the spectral line characteristics of oxygen A and B absorption bands. For spectral channels in the shoulder of the absorption band, the Monte Carlo method is used to analyze the influence of the position and the number of spectral channels and degree of fitting polynomial on the accuracy of non-absorption baseline fitting according to the sum of error squares and the correlation between the non-absorption baseline and the ideal baseline. For the spectral channels in the absorption band, the influence of the absorption rates in different spectral channels on the measurement range and the ranging accuracy is analyzed. Considering the real-time and precision of the system comprehensively, we find that each shoulder of the A absorption band has one spectral channel, and single shoulder of the B absorption band has two spectral channels. The location of spectral channels all should be chosen at the end of each shoulder near the absorption band. The spectral channels in the absorption band can be flexibly selected according to the requirements on the measurement rang and ranging accuracy. Therefore, if the complete spectral curve in the ranging waveband cannot be collected in one-time exposition, the monocular multispectral passive ranging system, using the fewer spectral channels and the simplest linear fitting method, will not only ensure the ranging accuracy of the system, but also reduce the time period of filter replacement and software calculation, and thus enhance the real-time performance of data acquisition and calculation.