Title: The method to capture the form of opera performance based on machine vision

Authors: Maoyuan Yin

Addresses: School of Music and Dance, Mudanjiang Normal University, Mudanjiang 157011, China

Abstract: In order to overcome the problems of traditional opera performance form capture methods, such as long denoising time and low capture accuracy, this paper proposes a new opera performance form capture method based on machine vision. The wavelet domain and gradient domain are obtained by multi-step local Wiener filtering in wavelet domain. The parameters of wavelet-based Wiener filter are modified to obtain a new wavelet threshold function. The grey images of different components of different opera performance form images are extracted for comparison, and the best component image is selected for threshold processing according to different opera performance forms. At the same time, image features are extracted in the form of binary image to realise the opera performance form capture based on machine vision. Experimental results show that the proposed method can quickly and accurately capture opera performance form, and the highest capture efficiency can reach 99.55%.

Keywords: machine vision; opera performance; form capture; noise variance; binary image.

DOI: 10.1504/IJART.2021.115767

International Journal of Arts and Technology, 2021 Vol.13 No.1, pp.30 - 44

Received: 06 Jul 2020
Accepted: 09 Nov 2020
Published online: 21 Jun 2021 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article