摘要
Abstract
[Research purpose]This study aims to explore the application of Digital Twin technology in resilient city development,with a particular focus on the integration of anomalous sound recognition technology in public spaces.The objective is to enhance urban capabilities in detecting and responding to public safety incidents,as well as to effectively address risks posed by natural disasters and human-induced threats.[Research method]By establishing a high-fidelity digital twin model of the urban environment,dynamic monitoring,simulation,and predictive analytics can be enabled,thereby supporting the development of resilient cities.Through a systematic literature review,this study comprehensively examines algorithms for abnormal sound recognition,sound source localization techniques,and representative application scenarios.Furthermore,from both technological and practical perspectives,it assesses the key challenges in integrating abnormal sound recognition within digital twin systems,including data quality,algorithmic robustness,hardware resource limitations,and privacy protection concerns.[Research conclusion]The integration of digital twin technology and anomalous sound recognition holds significant potential for enhancing urban resilience,particularly by strengthening the city's capacity for rapid detection and precise response to sudden public safety incidents.However,the deployment of abnormal sound recognition in public spaces continues to face four key challenges:insufficient quality of effective data,limited algorithmic robustness,computational resource constraints,and heightened difficulties in ensuring privacy protection.Future technological advancements should focus on multi-modal sensory fusion,optimization of edge computing,and federated learning,among other approaches,to enable the development of urban resilience systems characterized by high efficiency,advanced intelligence,and strong privacy safeguards.关键词
韧性城市/数字孪生/公共空间/异音识别/声源定位Key words
resilient cities/digital twin/public spaces/anomalous sound recognition/sound source localization分类
安全科学
