火力与指挥控制2016，Vol.41Issue(5)：66-70,5.
基于深度卷积神经网络的蛇形机动航迹图像识别
Deep Convolutional Neural Networks for the Image Recognition of“S-Maneuver”Target
摘要
Abstract
To improve the interception ability of anti-aircraft weapons and solve the existing robustness problem in the “S-maneuver”target recognition algorithm,the tracks coordinate data into images are transformed and then took advantage of deep convolutional neural networks to recognize it. The feasible method to solve the maneuver range inconformity problem is proposed which existed as the coordinate data transformed into images. The suitable deep convolutional architecture and network parameters have been identified after plenty of experiments based on CAFFE software platform. It is proved to be an efficient method to improve the robustness of target“S-maneuver”recognition.关键词
蛇形机动/图像识别/深度卷积神经网络/CAFFEKey words
S-maneuver/image recognition/deep convolutional neural networks/CAFFE分类
信息技术与安全科学引用本文复制引用
郑昌艳,梅卫,王刚..基于深度卷积神经网络的蛇形机动航迹图像识别[J].火力与指挥控制,2016,41(5):66-70,5.基金项目
国防“十一五”预研基金（40405020204）；国防“十二五”预研基金资助项目 （）
