Real-Time Imaging Technology Can Help to Prevent Dust Explosions

Researchers at Purdue University have built an imaging application to detect suspended dust concentrations that pose a risk of exploding in agricultural, powder-handling, and manufacturing settings.

The application establishes a relationship between the light extinction coefficient and suspended dust concentration. It uses a camera or video recording device to image and determine suspended dust concentration in a daylight environment and to distinguish it from normal background noise. After capturing images of the suspended dust cloud, the application analyzes the light extinction coefficient to measure the suspended dust concentration. It uses a calibration process to eliminate possible effects from changing ambient light conditions and camera noise.

Researchers at Purdue University have developed an image- and video-based application using OpenCV algorithms that detect potentially explosive suspended dust concentration. The app uses a camera or a video recording device to image and determine suspended dust, and to accurately distinguish it from normal background noise. Courtesy of Kingsly Ambrose/Purdue University.

The researchers tested suspended dust concentrations ranging between 10 and 90 kg/m³. They used cornstarch, corn dust, and sawdust as test materials. They found that the light extinction coefficient correlated with the suspended dust concentration. The mass extinction coefficient of the three sample materials was in the range of 0.03 to 0.04. Professor Kingsly Ambrose said that the application successfully recognized 95% of sawdust and 93% of cornstarch particulates in the air.

Current technology for detecting dust levels is expensive, is difficult to install in a workspace, and separates dust matter into multiple filters that must be weighed and further manipulated for analysis, Ambrose said. “This technology is unique because it is easy to use without extended training, location-independent, and does not require permanent installations,” he said.

Ambrose and the team worked with the Purdue Research Foundation Office of Technology Commercialization to patent the technology. They are looking to license it and are seeking collaborators for further development.

The research was published in the Journal of Loss Prevention in the Process Industries (www.doi.org/10.1016/j.jlp.2020.104242).