A virtual perception method for urban noise: The calculation of noise annoyance threshold and facial emotion expression in the virtual noise scene

With the development of immersive virtual technology and GIS, the virtual perception for noise annoyance will lead into new research issues. The aim of this study is to construct a methodology to realize virtual noise perception. There are three aspects: calculating method for annoyance thresholds in different noise control zones, facial emotion expression generating method, and linking method between subjective response and emotion. A face-to-face social survey has been carried out among 7483 participants in three different noise control zones. The noise annoyance is assessed by five-graded scale. With stratified random sampling results of survey as input data, the analysis hierarchy process is adopted to calculate noise annoyance threshold. Then the emotion is determined by excessive level and linkage between five-graded semantic scale and emotion. Finally, the facial emotion expression is used to realize the visualized expression and virtual perception of noise annoyance. The key technologies are the emotion determination method and the facial expression generating with free-form deformation technique. To illustrate the application of the proposed methodology, the method has been applied to a local district in Qingdao, China. Comparing the regulatory standards, results of calculated threshold value for noise control zones are more closely aligned with the people's real subjective responses. At the same time, the facial emotion expression provides a visualization display to achieve people's emotional reactions under noise pollution. Consequently, the proposed method can provide a reference to the competent authorities to make better action plans to prevent or reduce annoyance. (C) 2015 Elsevier Ltd. All rights reserved.