Vision-Based Measurement of Dust Concentration by Image Transmission

Abstract

A vision-based measurement (VBM) of dust concentration with high accuracy can be used to analyze dust pollution and reveal the distribution law of dust concentration in a production workshop. In conventional VBM, grayscale and morphological features of a dust image are commonly used. However, their measurement accuracy is limited by the atmospheric light scattering effect of surrounding light and the occlusion effect of dust particles. Thus, a high accuracy VBM system is investigated that is based on the relationship between dust concentration and image transmission (RCT) model. Meanwhile, a calculation method for image transmission that utilizes the atmospheric light scattering effect and the occlusion effect of dust particles is proposed in this paper. The experimental results show that in the proposed system, the relative error is ±3.0%, the relative uncertainty is 2.05%, the correlation is 0.99, the measurement range is 0.5-1000 mg/m <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> , and the measurement period is 2 s. The RCT tends to be a negative correlation, which is a quadratic polynomial. The impact of the atmospheric light scattering effect and the occlusion effect of dust particles on measurement accuracy can be eliminated by utilizing image transmission. The proposed system can be used as an alternative measurement of dust concentration in the workshop.