红外光谱信号预处理的新方法

要实现快速、准确地检测远距离污染气体红外光谱，对光谱进行预处理，消除高频噪声以及低频扰动，提取光谱特征信息是其中重要的一环。提出了一种对光谱信号进行预处理改进的高低通综合滤波法。该方法采用平滑滤波法去除高频噪声，采用阈值拟合技术处理低频扰动。阈值拟合技术的应用使整个光谱预处理过程完全自动化。实验结果表明，高低通综合滤波法可以使光谱预处理问题得到有效解决。该方法具有自适应功能，操作简便，运行可靠。

New method of preprocessing infrared spectral signals

It′s one of the most important phases to speedily and accurately detect the remote pollutant gas infrared spectrum that removing the high frequency noise and the low frequency disturbance to preprocess the spectrum for extracting characteristic information. A kind of improved high-low pass synthetic filter method for spectrum preprocessing is introduced. The high frequency noise is processed with moving average and the low frequency disturbance is processed by the threshold fitting technique,which makes the preprocessing for spectral signals automated. Experiments show that the improved high-low pass synthetic filter method can resolve the problem of preprocessing spectrum effectively. This self-adaptive method can be manipulated conveniently and run dependably.