Infrared Aircraft Anti-interference Recognition Based on Feature Enhancement CenterNet

Infrared imaging guided air to air missile is a typical infrared imaging guided weapon system. In the process of attacking targets such as fighter and helicopter, when the target discovers that it has been intercepted, it will release interference to induce the missile to deviate from its trajectory, which brings great difficulties to the missile's successful destruction of the target. By utilizing the powerful learning and representation capabilities of deep learning networks for target features, an infrared aircraft anti-interference recognition algorithm based on feature enhancement CenterNet is proposed. In order to enhance infrared aircraft features, a channel feature enhancement module is constructed. This paper applies this module to different scale feature layers of the feature extraction network and performs feature fusion through feature pyramid network. In order to reduce the impact of background and interference on aircraft and place the learning focus of feature extraction network on aircraft and its neighborhood, a mixed attention mechanism is incorporated into the backbone network. The experimental results show that the proposed method can effectively achieve anti-interference recognition of infrared aircraft.