基于多模复合探测方法的无人艇目标识别研究
Study on target recognition of USV based on multi-mode composite detection method
周昇辉 1武军安 1郭锐1
作者信息
- 1. 南京理工大学智能弹药技术国防重点学科实验室,江苏南京 210094
- 折叠
摘要
随着水面无人艇技术的不断发展和应用,其对于舰船的威胁程度也日益加剧.以末敏弹打击水面无人艇为背景,为了提升多元激光/红外/毫米波探测器对水面小型目标的识别性能,提出 了 一 种基于多通道卷积神经网络(Multi-Channel Convolutional Neural Network,MCC-NN)和极端梯度提升决策树(Extreme Gradient Boosting,XGBoost)的复合探测信号识别方法MCCNN-XGB,同时构建了单通道CNN识别网络与基于人工特征提取的XGBoost识别算法作为对照,最终通过水面目标无人机载探测试验数据对上述三种模型的 目标识别性能进行评估与对比.测试结果表明,基于MCCNN-XGB的识别算法表现最佳,测试准确率达到了 97.26％.本文所提出的识别方法能够有效进行复合探测信号的特征提取,并且能够降低误识别率与漏识别率,具有较好的识别效果.
Abstract
With the continuous development and application of USV technology,its threat to ships is increasing.In or-der to improve the recognition performance of multi-component laser/infrared/millimeter wave detector on small sur-face targets,a composite detection signal recognition method MCCNN-XGB based on multi-channel convolutional neu-ral network(Multi-Channel Convolutional Neural Network,MCCNN)and extreme gradient lifting decision tree(Ex-treme Gradient Boosting,XGBoost)is proposed.At the same time,a single channel CNN recognition network and XG-Boost recognition algorithm based on artificial feature extraction are constructed as a comparison.Then,the target rec-ognition performance of the above three models is evaluated and compared through the test data of UAV mount USV target.The test results show that the recognition algorithm based on MCCNN-XGB performs the best,with a test accu-racy of 97.26％.The recognition method proposed in this paper can effectively extract the features of the complex de-tection signal,and can reduce the false recognition rate and missing recognition rate,which has a good recognition effect.
关键词
末敏弹/水面无人艇/复合探测/目标识别/卷积神经网络/机器学习Key words
terminal sensitive projectile/USV/multiple detection/object identification/CNN/machine learning引用本文复制引用
基金项目
面向成像式灵巧弹药的DCNN轻量化研究项目()
高动态导航技术北京市重点实验室开放基金任务书项目()
出版年
2024
NETL
NSTL
万方数据