Deep Learning based Asbestos Fiber Detection

Soma Biswas, Debmalya Biswas

Abstract: Airborne respirable fibers, such as asbestos are hazardous to health. Occupational health and safety guidelines and laws require detection and identification of all the asbestos containing materials. However, detection and identification of asbestos fibers is a complex, time-consuming and expensive process. In this work, we present a Deep Learning based Se-mantic Segmentation model that is able to automate the asbestos analysis process, reducing the turnaround time from hours to minutes. The proposed deep neural network provides end-to-end automation of the analysis process, starting with transforming the input Scanning Electron Microscope (SEM) images, to identifying and counting the number of fibers in the image, to masking the identified fiber regions and re-arranging for efficient processing by Energy Dispersive Spectroscopy (EDS). Finally, we provide implementation details of a U-Net based Semantic Segmentation model that is able to detect and count asbestos fibers (air sample) in SEM images with up to 95% accuracy.