A method to reduce open optical path noise interference in two-dimensional gas detection of farmland

Abstract

The two-dimensional distribution method based on the laser spectrum is widely used in the reconstruction of two-dimensional temperature fields or concentration fields. However, the existing methods are all based on small-scale measurements. In large-scale spatial detection of farmland, the detection accuracy cannot reach the expected result due to the long optical path and the noise of various frequencies brought by environmental factors such as vibration, temperature, and other interfering gases. Aiming at the above-mentioned problems, this paper proposes a method to reduce open optical path noise interference in two-dimensional gas detection of farmland to improve detection accuracy. The signal denoising approach combines variational modal decomposition with the SG filtering algorithm to decompose, smooth out, and rebuild the detection signal. In addition, we built a two-dimensional detection system based on an open optical path. Throughout experimental validation, the proposed method outperformed the traditional method by 24% and 19%, respectively, in terms of the fitting degree and fitting effect of the gas absorption curve.