Command and Control Engineering College,Army Engineering University of PLA
Natural Science Foundation of Jiangsu Province (Project Number: BK20200581); National Natural Science Foundation of China (61806220); Frontier Science and Technology Innovation Program of Army Engineering University of PLA
In the complex battlefield environment, there are often a lot of similar military targets, such as: different types of armored vehicles, aircraft and warships. The task of recognizing such targets called fine-grained recognition in the field of computer vision, and becomes more challenging than a normal classification task because of the subtle differences between the targets. At the same time, with the continuous development of military equipment, similar military targets often have extremely different combat capabilities. Therefore, how to recognize the fine-grained military targets is of great significance to help commanders grasp battlefield information more accurately and assist their command. In view of this problem, this paper introduces the development status of fine-grained recogniztion technology, and analyzes the application prospect and challenge of fine-grained recogniztion of military targets combined with the practical application scenarios of military target identification, and gives the possible solutions.