Spatiotemporal fusion strategies in multi-modal sensor target detection and tracking

Although significant progress has been made in the research of object detection and tracking algorithms in recent years, there are still some unresolved issues that affect the practicality of the algorithms. This article mainly studies the optimal and suboptimal filtering algorithms used to process various linear and nonlinear models in target tracking, and then uses different filtering algorithms to filter and estimate the motion states of multiple targets, aiming to improve the accuracy of target tracking, and analyses and compares the advantages and disadvantages of these algorithms and their application scenarios. The experimental results show that the target tracking accuracy of the multi-sensor algorithm studied in this paper has been improved by 39.1%. While maintaining the same accuracy of 5.976 times, the time limit of the image information fusion algorithm studied in this paper is 142.187 ms, and the efficiency of the optimised algorithm is 142.187 ms.