通过特征融合可获取全面的目标特征信息，利于提升目标识别的稳定性，为此设计基于特征融合的无人船目标识别系统。利用无人船搭载红外热成像仪与可见光摄像头，采集目标红外与可见光图像；通过处理器和可编程逻辑控制器，设计特征提取模块，用于提取红外与可见光图像的无人船目标特征；特征融合模块利用典型相关分析理论，融合红外与可见光图像的无人船目标特征；目标识别模块通过径向基函数网络，结合特征融合结果，输出无人船目标识别结果。实验结果证明，该系统可有效采集无人船目标的红外与可见光图像，完成特征提取；该系统具备较优的特征融合效果，并精准实现无人船目标识别。
Through feature fusion, comprehensive target feature information can be obtained, which is conducive to improving the stability of target recognition. Therefore, an unmanned ship target recognition system based on feature fusion is designed. Infrared and visible images of the target are collected by the unmanned ship equipped with infrared thermal imager and visible light camera. Based on the processor and programmable logic controller, a feature extraction module is designed to extract the target features of the unmanned ship in infrared and visible images. The feature fusion module uses the theory of canonical correlation analysis to fuse the target features of the unmanned ship in infrared and visible images. The target recognition module outputs the target recognition result of unmanned ship through radial basis function network and feature fusion result. Experiments show that the system can effectively collect infrared and visible images of unmanned ship targets and complete feature extraction. The system has better feature fusion effect and accurately realizes the target recognition of unmanned ship.
2024,46(12): 174-177 收稿日期：2023-12-06
DOI：10.3404/j.issn.1672-7649.2024.12.031
分类号：TP391.4
基金项目：江苏省高等学校基础科学（自然科学）研究重大项目（23KJA580002）；2022江苏省“青蓝工程”优秀教学团队（苏教师函（2022）29号）
作者简介：颜悦(1985-),女,硕士,讲师,研究方向为人工智能。
参考文献：
[1] 初庆栋, 尹羿博, 龚小旋, 等. 基于双偶极向量场的欠驱动无人船目标跟踪制导方法[J]. 中国舰船研究, 2022, 17(4): 32-37.
CHU Qingdong, YIN Yibo, GONG Xiaoxuan, et al. Method for target tracking and guidance of under-actuated unmanned surface vehicle based on two-dimensional dipolar vector field[J]. Chinese Journal of Ship Research, 2022, 17(4): 32-37.
[2] 何志强, 专祥涛, 梁杰, 等. 基于区域人工势场的无人船目标跟踪策略研究[J]. 中国造船, 2023, 64(1): 236-245.
HE Zhiqiang, ZHUAN Xiangtao, LIANG Jie, et al. Research on target tracking strategy of unmanned surface vessel based on regional artificial potential field[J]. Shipbuilding of China, 2023, 64(1): 236-245.
[3] 倪桦, 关巍, 张显库. 基于激光雷达与摄像头的无人船目标感知与测距[J]. 船舶工程, 2022, 44(9): 107-113.
NI Hua, GUAN Wei, ZHANG Xianku. Target perception and location of unmanned ship based on lidar and camera[J]. Ship Engineering, 2022, 44(9): 107-113.
[4] 王哲昊, 简涛, 黄晓冬, 等. 基于角域特征PSO的海面目标HRRP识别方法[J]. 系统工程与电子技术, 2023, 45(6): 1642-1650.
WANG Zhehao, JIAN Tao, HUANG Xiaodong, et al. HRRP recognition method for sea surface targets based on angular domain feature PSO[J]. Systems Engineering and Electronics, 2023, 45(6): 1642-1650.
[5] 刘可佳, 马荣生, 唐子木, 等. 采用优化卷积神经网络的红外目标识别系统[J]. 光学精密工程, 2021, 29(4): 822-831.
LIU Kejia, MA Rongsheng, TANG Zimu, et al. Design of infrared target recognition system with optimized convolutional neural network[J]. Optics and Precision Engineering, 2021, 29(4): 822-831.
[6] 孙祎芸, 樊臻, 董山玲, 等. 基于双流对称特征融合网络模型的海洋船舶目标识别[J]. 控制理论与应用, 2022, 39(11): 2009-2018.
SUN Yiyun, FAN Zhen, DONG Shanling, et al. Marine ship target recognition using two-stream symmetric feature fusion convolutional neural network[J]. Control Theory & Applications, 2022, 39(11): 2009-2018.
