An AI based smart-phone system for asbestos identification

Michael Rolfe; Samantha Hayes; Meaghan Smith; Matthew Owen; Michael Spruth; Chris McCarthy; Abdur Forkan; Abhik Banerjee; Rosalie K Hocking

doi:10.1016/j.jhazmat.2023.132853

An AI based smart-phone system for asbestos identification

J Hazard Mater. 2024 Feb 5:463:132853. doi: 10.1016/j.jhazmat.2023.132853. Epub 2023 Oct 27.

Authors

Michael Rolfe¹, Samantha Hayes², Meaghan Smith¹, Matthew Owen³, Michael Spruth², Chris McCarthy¹, Abdur Forkan¹, Abhik Banerjee¹, Rosalie K Hocking⁴

Affiliations

¹ Department of Chemistry and Biotechnology and Department of Computing Technologies, School of Science, Computing and Engineering Technologies, Swinburne University of Technology Melbourne, VIC 3122, Australia.
² Agon Environmental Pty, Ltd 63-85 Turner Street, Port Melbourne, VIC 3207, Australia.
³ Identifibre Pty Ltd., 67 Atherton Road, Oakleigh, VIC 3166, Australia.
⁴ Department of Chemistry and Biotechnology and Department of Computing Technologies, School of Science, Computing and Engineering Technologies, Swinburne University of Technology Melbourne, VIC 3122, Australia. Electronic address: rhocking@swin.edu.au.

PMID: 37918071
DOI: 10.1016/j.jhazmat.2023.132853

Abstract

Asbestos identification is a complex environmental and economic challenge. Typical commercial identification of asbestos involves sending samples to a laboratory where someone learned in the field uses light microscopy and specialized mounting to identify the morphologically distinct signatures of Asbestos. In this work we investigate the use of a portable (30x) microscope which works with a smart phone camera to develop an image recognition system. 7328 images from over 1000 distinct samples of cement sheet from Melbourne, Australia were used to train a phone-based image recognition system for Asbestos identification. Three common CNN's were tested ResNet101, InceptionV3 and VGG_16 with ResNet101 achieving the best result. The distinctiveness of Asbestos was found to be identified correctly 90% of the time using a phone-based system and no specialized mounting. The image recognition system was trained with ResNet101 a convolutional neural network deep learning model which weights layers with a residual function. Resulting in an accuracy of 98.46% and loss of 3.8% ResNet101 was found to produce a more accurate model for this use-case than other deep learning neural networks.

Keywords: Asbestos; Constitutional Neutral Network; Hazardous Materials Identification; Image Recognition.