Research on Digital Twin Construction and Immersive Teaching Application of Luoyang Cultural and Tourism Heritage

Under the background of parallel promotion of digital teaching and cultural heritage activation, although Luoyang cultural tourism heritage has a good digital acquisition foundation, there are still some problems such as scene reconstruction emphasizing display, insufficient semantic organization, and weak teaching adaptation. Therefore, this paper constructs a digital twin scene model of Luoyang cultural tourism heritage, and proposes a virtual-real fusion interactive application method for immersive teaching. In the scene construction task, the Longmen Grottoes, the White Horse Temple and the ruins of Luoyang City in the Sui and Tang Dynasties are selected as the objects, and the laser point cloud, oblique photography images, panoramic sequences and historical text data are fused. In the teaching application task, the classroom interaction data and training process data are introduced to comprehensively evaluate the system effect. The results show that the geometric construction accuracy of the proposed method in the three types of heritage scenes reaches 89.67%, 93.21%and 91.34%respectively, and the comprehensive geometric construction accuracy and semantic anchor matching rate reach 95.84%and 94.36%respectively under mixed data sources. In the immersive teaching application, the knowledge test accuracy and task completion rate reached 91.48%and 89.36%respectively, and the average interactive response time was 96.14 ms. The experimental results show that the proposed method improves the construction quality and teaching application effect of Luoyang cultural tourism heritage digital scene through the collaborative design of multi-source data fusion, semantic scene modeling and dynamic interactive scheduling.