论文部分内容阅读
利用红外遥测光谱仪远距离快速、准确地探测污染气体并给出定性鉴别结果,必须对遥测光谱进行预处理,消除高频噪声和低频基线的干扰,提取出污染气体的特征光谱信号。针对现有方法的不足,提出采用具有自适应特性的EMD方法,对光谱信号进行无参数分解,提取出高频噪声与低频基线,实现了红外遥测光谱的预处理。经过该方法处理,光谱信号全局评估系数RMS1平均值达到0.141,局部评估系数RMS2平均值达到0.182,综合评估系数RMS*平均值达到0.026,明显优于小波方法。实验结果表明,EMD方法用于红外遥测光谱信号去噪与基线校正,算法简单,运行可靠,可使问题得到有效解决。
Using infrared telemetry spectrometer to detect contaminated gas quickly and accurately and give qualitative identification results, the telemetry spectrum must be preprocessed to eliminate the interference of high-frequency noise and low-frequency baseline and to extract the characteristic spectral signal of contaminated gas. Aiming at the shortcomings of the existing methods, this paper proposes EMD method with adaptive characteristics to decompose the spectrum signal without parameters and extract high frequency noise and low frequency baseline to achieve the pretreatment of infrared telemetry spectrum. After this method, the average value of spectral signal global evaluation coefficient RMS1 reaches 0.141, the local evaluation coefficient RMS2 average value reaches 0.182, and the comprehensive evaluation coefficient RMS * average value reaches 0.026, which is obviously better than the wavelet method. Experimental results show that the EMD method is suitable for denoising and baseline correction of infrared telemetry signals. The algorithm is simple and reliable, which can effectively solve the problem.