Practical Exploration of Digital Twin Technology to Construct a Teaching Environment Combining Virtual and Real in Civic and Political Education

Abstract

The rapid development of information technology has made digital twin technology a key driver of educational transformation. This paper combines high-performance sensors and other hardware with feature extraction algorithms and other technologies to construct an online immersive teaching system that integrates reality and virtuality, thereby achieving the digitization and virtualization of ideological and political education. The TSDF algorithm, Neural Recon algorithm, and Marching Cube algorithm are used to process spatial data at high speed, and ThingJS and GIS technology are employed to convert between virtual and real coordinates, thereby enhancing the interactive performance of the model. Research findings indicate that the model's accuracy, completeness, and average value metrics reach 0.3013 mm, 0.1718 mm, and 0.4011 mm, respectively, with iteration loss values ranging from 2.50 to 9.98, outperforming comparison models in reconstruction performance. When the model is used to assist ideological and political education, the experimental class demonstrates significantly higher ideological and political proficiency across five dimensions compared to the control class (p < 0.01).