Developing and validating an image processing algorithm for evaluating gravel road dust

Daily traffic on arid gravel roads can easily generate dust. Dust emitted from gravel roads creates several problems such as aggravated asthma, breathing difficulties, reducing crop yields, and even death. Therefore, local agencies tend to track the dust amounts on the gravel roads in order to maintain them in good conditions. An accurate detection of dust amounts is very crucial in Gravel Roads Management System (GRMS). Data collection is considered as one of the main challenges facing local agencies due to budget constraints. This paper establishes a novel method for an automatic recognition of dust amounts on gravel roads. This algorithm, “Simple Dust Classification Algorithm”, uses images taken from smartphone application, “Road roid”, to classify dust amounts by using proprietary Digital Image Processing algorithms. A Dustometer device was used to validate the proposed algorithm. Dustometer measurements, supported by statistical analysis, demonstrate that the proposed algorithm achieves an outstanding dust amount classification accuracy. Hence, the proposed algorithm is a promising alternative to assist local agencies in data collection and maintenance planning.