This research aims to explore the method of combining digital twins (DTs) with Convolutional Neural Network (CNN) algorithms to analyze the attraction of museum historical and cultural exhibits to achieve intelligent and digital development of museum exhibitions. Firstly, the DTs technology is used to digitally model the museum's historical and cultural exhibits, realizing virtual exhibit display and interaction. Then, the Mini_Xception network is proposed to improve the CNN algorithm, which is combined with the ResNet algorithm to construct a human facial emotion recognition model. Finally, the proposed model is used to accurately predict the attraction of museum historical and cultural DTs exhibits by recognizing people's facial expressions when observing them. Comparative experimental results show that the proposed recognition method can greatly improve accuracy and scalability. Compared with traditional recognition methods, the recognition accuracy can be improved by 5.53%, and 2.71s can reduce the model's data transmission delay. The enhanced scalability of the recognition type can also meet the real-time interaction requirements in a shorter time. This research has important reference value for the digital and intelligent development of museum historical and cultural exhibitions.
Abstract
This research aims to explore the method of combining digital twins (DTs) with Convolutional Neural Network (CNN) algorithms to analyze the attraction of museum historical and cultural exhibits to achieve intelligent and digital development [...]