Title: Visual perception-based human-computer interaction information classification method for intelligent products

Authors: Liting Zhou; Xuan Li; Minmin Guo

Addresses: School of Art Design, Shandong Youth University of Political Science, Jinan, Shandong, China ' School of Art Design, Shandong Youth University of Political Science, Jinan, Shandong, China ' College of Art and Design, Zaozhuang University, Zaozhuang, Shandong, China

Abstract: This paper proposes a new intelligent product human-computer interaction information classification method based on visual perception. Design smart product human-computer interaction information collection device to realise rapid and accurate collection of smart product human-computer interaction information, and fusion processing of the collected information. The ISA model is built according to the principle of visual perception, and the model is optimised by the gradient descent method. The optimised model is used to extract the information attribute characteristics, and the intelligent product human-computer interaction information classification is carried out according to the information attribute characteristics. The experimental results show that the accuracy of information classification of this method is always above 94.7%, and the average classification time is 0.53 s, which verifies the superiority of the method.

Keywords: visual perception; intelligent products; man-machine interaction; information classification; ISA model.

DOI: 10.1504/IJPD.2023.129308

International Journal of Product Development, 2023 Vol.27 No.1/2, pp.28 - 40

Received: 07 May 2021
Accepted: 24 Nov 2021
Published online: 06 Mar 2023 *