Spatiotemporal analysis and visualization system design for moving objects based on trajectory data mining

Analysis of moving objects is a significant problem in various areas like transportation. However, it has low computational efficiency as well as low accuracy. In order to improve this problem, this study proposed a moving object spatiotemporal analysis model by using trajectory data analysis. The model first applies Kalman filtering for noise reduction. Then it uses the dynamic time warping algorithm and the hierarchical clustering algorithm to calculate similarity and conduct clustering analysis. After that, the model carries out spatiotemporal analysis of moving objects through convolutional neural network, Bi-directional long short-term memory, and transformer-extra long. Moreover, the system is able to visualize the result of the spatiotemporal analysis. From the experiments on the practical application of the system, it is shown that when the time unit is set to 300 s and 600 s, the accuracy of the proposed system reaches 93.12% and 93.33%. These two figures are higher than those of the comparison models, suggesting that the system has stronger spatiotemporal analysis ability. In addition, the speed of the system’s calculation is up to 95.17 fps, which is higher than that of the comparison models. The results of the experiments show that the accuracy and computational efficiency of the proposed model are higher. The proposed model can efficiently handle spatiotemporal analysis of moving objects and offer important technology support in intelligent applications of urban traffic management, dynamic resource allocation, and prevention of public security.