Battlefield Image Situational Awareness Application Based on Deep Learning
Hui Peng,Yifan Zhang,Bin Song,Sen Yang
DOI: https://doi.org/10.1109/MIS.2019.2953685
IF: 6.744
2020-01-01
IEEE Intelligent Systems
Abstract:With the rapid development of information technology, it has become an important topic to construct a situational awareness system that can independently mine data and information as well as perceive environmental situations by using deep learning. First, this article introduced the structure of convolutional neural networks (CNN) and You Only Look Once (YOLO) model. Then, it analyzed the structure and function of battlefield situational awareness system, and concluded that: in the whole situational awareness system, the discovery, category, and location analysis of situational elements, namely object target, is the foundation and key to realize the function. On this basis, this article establishes a battlefield situational awareness model based on the YOLO model. Finally, five common objects on the battlefield (helicopter gunship, missile, tank, soldier and gun) are classified and located, respectively. The YOLO model based on CNN is used to process the input image, and then the position, category, and corresponding confidence probability of all objects in the image are obtained directly, which realizes end-to-end learning, greatly improves the speed of target detection, and lays a foundation for assessing the battlefield situation.
Engineering,Computer Science