Objectives Earthquake disaster has become one of the natural disasters that pose a great threat to human society, and it puts forward high requirements for the efficiency and accuracy of emergency rescue. Augmented reality (AR) technology can provide more intuitive decision-making support for rescue personnel by integrating virtual information into the real scene. However, the existing AR scene modeling technologies face challenges in the complex environment of earthquake disaster.

Methods This paper proposes an accurate modelling method for earthquake disaster AR scenes with structural semantic feature constraints. First, the characteristics of earthquake disaster scenes are analyzed, and the semantic feature library of earthquake disaster scene structure is constructed. Second, the virtual-real feature extraction of earthquake disasters with structural semantic constraints is proposed to improve the accuracy of virtual-real image feature extraction. Then, the virtual-real modelling of the earthquake disaster AR scene is carried out based on the results of virtual-real feature extraction. Finally, the damaged buildings are selected as the experimental case area for experimental analysis.

Results Experimental results indicate that the proposed method achieves an F1 score of 90% for earthquake disaster feature extraction based on structural semantic constraints. The optimized AR scene registration error has been reduced by 80% compared to the direct modeling method.

Conclusions The findings demonstrate that the proposed method can effectively extract and match virtual-real features, leading to accurate AR scene modelling for earthquake disaster scenarios. At the same time, the proposed method has important application potential for improving the accuracy and reliability of AR applied to disaster emergency response, and can assist on-site rescue personnel to obtain more accurate disaster information.