摘要
[目的]旨在解决由于小目标船舶特征不显著导致网络模型识别率低的问题,提出基于图像与运动特征融合方法。[方法]在小目标船舶图像特征不明显的情况下,融合运动特征信息,丰富小型船舶的特征表达,同时提出混合注意力模型。在数据驱动条件下加入船舶目标先验信息,提高模型对关键特征的感知和利用能力。[结果]在720P分辨率图像中能实现10×10像素的小目标船舶识别,识别距离达到4 km范围,实现了广域船舶识别与定位功能。[结论]改进后的目标识别网络既具备像素级别的小目标船舶检测能力,同时又具备对环境噪声的抗干扰能力,突破了网络模型对小目标船舶识别率低的瓶颈。
[Objective]This paper addresses a problem affecting small target ship detection in which network models have a low recognition rate caused by the insignificance of features in small target ships.[Methods]A fusion method based on the integration of image and motion features is proposed to enrich the feature representation of small ships in scenarios where the features of small target ship images are not prominent.Additionally,a hybrid attention model incorporates the prior information of ship targets under data-driven conditions to enhance the model's perception and utilization of key features.[Results]The proposed method achieves the recognition of small target ships with a resolution of 720P at a distance of up to 4 kilometers,enabling wide-area ship recognition and localization functionality.[Conclusion]The improved target recognition network exhibits pixel-level small target detection capability while also demonstrating robustness against environmental noise interference,thereby overcoming the bottleneck of the low recognition rate of network models in small target ship detection.
作者
严荣慧
郭前
雷鸣
蔡雁翔
羊箭锋
YAN Ronghui;GUO Qian;LEI Ming;CAI Yanxiang;YANG Jianfeng(School of Intelligent Manufacturing and Smart Transportation,Suzhou City University,Suzhou 215006,China;School of Electronic and Information Engineering,Soochow University,Suzhou 215006,China)
出处
《中国舰船研究》
CSCD
北大核心
2024年第6期284-292,共9页
Chinese Journal of Ship Research
基金
江苏省高等学校自然科学研究面上资助项目(21KJD510006,23KJB510026)。
作者简介
严荣慧,女,1993年生,硕士,讲师。研究方向:模式识别与人工智能。E-mail:rhyan206@163.com;郭前,女,1997年生,博士,讲师。研究方向:模型预测控制。E-mail:qguo@szcu.edu.cn;雷鸣,男,1987年生,硕士,讲师。研究方向:机械动力学。E-mail:sudaleiming@163.com;通信作者:羊箭锋,男,1978年生,博士,副教授。研究方向:信号与信息处理。E-mail:jfyang@suda.edu.cn。
