Laundry to Laboratory: Automated Image Analysis for the Characterization of Fibrous Microplastics

Microfibers from clothes laundering are the predominant source of ocean microplastics. Despite this, existing research relies on outdated methods and is inconsistent. This study addresses the critical need for improved characterization of the prevalence and morphology of fibrous microplastics by introducing an automated image analysis approach. Utilizing digital or smartphone cameras, significant improvements are achieved in the characterization rate without losing accuracy, thus reducing variability and uncertainty. A novel Gaussian-offset threshold methodology for automatic microplastic segmentation demonstrates a 95% binary classification accuracy and a Matthews correlation coefficient of 0.87. Our methods were primarily evaluated using polyester microplastics from clothes washing produced at a concentration of 6280 fibers/g with a median length of 435 μm. Additional testing was conducted with fibers of varied makeup, width, and topological complexity. The practical application of this research was exemplified through a froth flotation study. Bubble flux was optimized for microplastic removal, and microplastic concentrations and length distributions were tracked over time. This work can be easily integrated into existing practices, significantly improves the labor-intensive nature of characterization, and ultimately contributes to a more standardized and reliable approach to understanding and mitigating fibrous microplastic pollution.