Abstract
Surface dust can be a source carrier of viruses, bacteria and air pollution which entail common health issues such as asthma attacks, chest tightness, wheezing and difficulty in breathing. Visually perceived, cleanliness is one measure of indoor air quality and is the subjective assessment of cleaning quality. The aim of this work is to use pattern recognition mediated through a mobile application to analyse and classify dust in households, in order to obtain useful information about the dust sources for the selection of appropriate countermeasures in view to improve air quality and better manage the cleaning of indoor surfaces. The dust type categorization in this work are pollen, rock and ash. This paper also explores the concept of transfer learning techniques and adopts it for small particle classification using CNN models by developing a surface dust application for android smartphones. The behaviour of InceptionV2, InceptionV3, ResNetV2, MobileNetV2 and MobileNetV3 as dust feature extractors were analysed based on their accuracy, precision, recall and F1-Score performance metrics. Results show that MobileNetV3 model is best suited as a dust feature extractor and rapid dust prediction with an accuracy of up to 92% and low-size storage of only 30 megabytes.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.