为准确划分远洋船舶类别，实现远洋船舶目标检测，保障海洋生态环境、海洋交通与国防安全，研究远洋船舶目标检测中图像分类识别方法。采用小波分析法，通过二维小波变换获取低频子图像、水平细节子图像、垂直细节子图像和对角细节子图像，将其作为远洋船舶图像特征；采用采用双向递归神经网络(BRNN)的深度学习方法构建分类器，将远洋船舶图像特征作为输入向量，通过递归神经网络将输入向量映射至输出向量，得到船舶目标图像分类结果。同时利用改进粒子群算法优化分类器中的权重与偏差参数，提升船舶目标图像分类精度。实验结果显示，所研究方法能够有效划分船舶目标图像类别，且划分精度高于97%。
In order to accurately classify ocean ships, realize the target detection of ocean ships, and ensure the protection of marine ecological environment, marine traffic and national defense security, the image classification and recognition methods in ocean ship target detection are studied. The low-frequency sub image, horizontal sub image, vertical sub image and diagonal sub image are obtained by two-dimensional wavelet transform, which are used as the image features of ocean going ships; The deep learning method of bidirectional recurrent neural network (brnn) is used to construct a classifier. The features of ocean going ship images are taken as input vectors, and the input vectors are mapped to the output vectors through the recurrent neural network to obtain the classification results of ship target images; At the same time, the improved particle swarm optimization algorithm is used to optimize the weight and deviation parameters in the classifier, so as to improve the classification accuracy of ship target images. Experimental results show that the proposed method can effectively classify ship target images, and the classification accuracy is higher than 97%.
2022,44(18): 177-180 收稿日期：2022-06-02
DOI：10.3404/j.issn.1672-7649.2022.18.037
分类号：TP391
基金项目：山西省自然科学基金资助项目(202103021224204)
作者简介：陈燕(1982-),女,博士,讲师,研究方向为信号与信息处理及图像处理
参考文献：
[1] 马啸, 邵利民, 金鑫, 等. 基于改进Mask R-CNN的可见光图像中舰船目标检测方法[J]. 北京理工大学学报, 2021, 41(7): 734–744
MA Xiao, SHAO Limin, JIN Xin, et al. Ship Target Detection in Optical Images Based on Improved Mask R-CNN[J]. Transactions of Beijing Institute of Technology, 2021, 41(7): 734–744
[2] 李健伟, 曲长文, 彭书娟. 基于级联CNN的SAR图像舰船目标检测算法[J]. 控制与决策, 2019, 34(10): 2191–2197
LI Jian-wei, QU Chang-wen, PENG Shu-juan. A ship detection method based on cascade CNN in SAR images[J]. Control and Decision, 2019, 34(10): 2191–2197
[3] 周慧, 严凤龙, 褚娜, 等. 基于特征金字塔模型的高分辨率遥感图像船舶目标检测[J]. 大连海事大学学报, 2019, 45(4): 131–138
ZHOU Hui, YAN Feng-long, CHU Na, et al. Ship target detection in high resolution remote sensing images based on characteristic pyramid model[J]. Journal of Dalian Maritime University, 2019, 45(4): 131–138
[4] 王昌安, 田金文. 生成对抗网络辅助学习的舰船目标精细识别[J]. 智能系统学报, 2020, 15(2): 296–301
WANG Chang'an, TIAN Jinwen. Fine-grained inshore ship recognition assisted by deep-learning generative adversarial networks[J]. CAAI Transactions on Intelligent Systems, 2020, 15(2): 296–301
[5] 陈彦彤, 李雨阳, 姚婷婷. 复杂海况下遥感图像舰船目标检测方法研究[J]. 激光与光电子学进展, 2019, 56(18): 152–160
CHEN Yantong, LI Yuyang, YAO Tingting. Ship Detection from Remote Sensing Image Under Complex Sea Conditions[J]. Laser & Optoelectronics Progress, 2019, 56(18): 152–160
[6] 杨龙, 苏娟, 黄华, 等. 一种基于深层次多尺度特征融合CNN的SAR图像舰船目标检测算法[J]. 光学学报, 2020, 40(2): 132–140
YANG Long, SU Juan, HUANG Hua, et al. SAR Ship Detection Based on Convolutional Neural Network with Deep Multiscale Feature Fusion[J]. Acta Optica Sinica, 2020, 40(2): 132–140
[7] 孙忠镇, 戴牧宸, 雷禹, 等. 基于级联网络的复杂大场景SAR图像舰船目标快速检测[J]. 信号处理, 2021, 37(6): 941–951
SUN Zhongzhen, DAI Muchen, LEI Yu, et al. Fast Detection of Ship Targets for Complex Large-scene SAR Images Based on a Cascade Network[J]. Journal of Signal Processing, 2021, 37(6): 941–951
