Intelligent Compensation Method of Infrared Temperature Measurement for Multiple Interference Factors
Dong Pan, Zhaohui Jiang, Yitian Li, Haoyang Yu, Weihua Gui
Abstract
Infrared thermography is a widely used noncontact temperature measurement method. However, the infrared temperature measurement result is susceptible to interference factors. To overcome the infrared temperature measurement errors caused by the simultaneous influence of multiple interference factors, this article takes the measuring distance and dust as specific interference factors and proposes an intelligent compensation method for the infrared temperature measurement errors caused by the measuring distance and dust. First, the influence of the measuring distance and dust is analyzed from the perspective of the infrared temperature measurement principle. Second, a weighted ensemble stacked denoising autoencoder (WE-SDAE) is proposed and used to construct an intelligent compensation model. Then, the measuring distance, dust transmittance, and measured temperature are used as the input of the compensation model to estimate the errors caused by the measuring distance and dust. Finally, the estimated error is used to compensate for the original infrared temperature measurement results under multifactor interference. Experimental results in two cases show that the proposed method is effective in compensating for the infrared temperature measurement errors caused by the measuring distance and dust, which is beneficial to ensure the accuracy of the infrared thermal imager in complex temperature measurement scenarios.