Moving Target Recognition in Noisy Environment Based on Sound Seismic Fusion Signal and PSO-SVM

Abstract

Recognition of moving targets (pedestrian and vehicle) by ground seismic signal has been a hot research topic in recent years. The methods proposed in previous studies mainly focus on seismic signal recognition in field environment. However, studies involving target recognition in noisy environment suffer from low recognition rate and slow recognition speed. To solve the problem of target recognition in noisy environment and to improve recognition performance, a method based on sound seismic fusion signal and particle swarm optimization support vector machine (PSO-SVM) is proposed. The method consists of three parts: signal acquisition, feature extraction and classification method. The method innovatively introduces sound seismic fusion signal as sampling data to improve the feature extraction efficiency of the target, and uses particle swarm optimization support vector machine to further improve the recognition accuracy. Compared with other methods, the fused signal can better collect the features of the target in noisy environment, and the optimized algorithm has the features of simple structure, high efficiency and high accuracy. After experimental verification, the recognition accuracy of the method for moving targets in noisy environment is 93.21%.