Artificial intelligence technology has a pivotal position in modern war fare.The precise identification of military targets in a complex environment wi ll help us seize the opportunity to defeat the enemy.Adhering to the important concept of sea air integrated joint operation,a lightweight and accurate milita ry target detection method is proposed in order to embed the target detection me thod into the microcomputers of various military units.By analyzing and combini ng the characteristics of soldiers,cars,tanks and their tracks,a lightweight network unit is designed.The network unit is used to form a backbone network wi th low computational complexity and accuracy,which is used to extract the targe t feature information,which improves the detection speed of the method.A MSCA (multi-scale context aggregation) module is designed to improve the high-dimen si onal performance of the backbone network.Separate extraction from low-dimensio n al features solves the problem of target occlusion.Experimental results show th at the accurate recognition rate of this method in the military target test set is 97.8%.Compared with the YOLOv4detection method,the detection accuracy is i mproved by 1.1%,and the running speed is increased by 5times,which can meet t he requirements of real-time operation of embedded devices.Through experiments , the proposed method can detect military targets in a variety of scenarios in re al time and accurately.