Non-contact identification of moisture content of fabric based on analysis of broadband acoustic signals using multiple-frequency air ultrasonic transducer system

Abstract

We fabricated a non-contact identification system employing multiple-frequency air ultrasonic transducers and a microphone capable of broadband measurement. This study aims to perform non-contact identification of the state of cloth using broadband acoustic analysis and machine learning. We conducted experiments to obtain basic data on the relationship between the moisture content of cloth and the frequency–amplitude characteristics. Using the proposed system, which combines high-resolution acoustic measurement and machine learning, we succeeded in noncontact identification of the moisture content of fabric. In addition, we verified the feasibility of this system in identifying whether the fabric material is cotton or polyester.