以提升舰船颜色特征的应用技术水平，研究三维动态舰船图像颜色特征自动提取和应用方法。通过相机采集三维动态舰船图像颜色特征后，先使用PCA算法获取三维动态舰船图像的颜色特征子空间，再通过K-means聚类算法得到三维动态舰船图像颜色特征，以该颜色特征作为基础，分别利用支持向量机算法和二阶常速模型实现舰船目标识别和航迹跟踪。实验结果表明，该方法可有效提取三维动态舰船图像颜色特征的R、G、B分量，提取舰船图像颜色特征能力较强。将提取到的三维动态舰船图像颜色特征，应用到舰船目标识别和航迹跟踪，可有效识别舰船和跟踪舰船航迹，应用效果较为显著。
To study the automatic extraction and application methods of color features in 3D dynamic ship images, and improve the application technology level of ship color features. After collecting the color features of the 3D dynamic ship image through the camera, the PCA algorithm is first used to obtain the color feature subspace of the 3D dynamic ship image. Then, the K-means clustering algorithm is used to obtain the color features of the 3D dynamic ship image. Based on this color feature, support vector machine algorithm and second-order constant velocity model are used to achieve ship target recognition and track tracking. The experimental results show that this method can effectively extract the R, G, and B components of the color features of 3D dynamic ship images, and has a strong ability to extract the color features of ship images. The extracted 3D dynamic ship image color features can be applied to ship target recognition and track tracking, and can effectively identify ships and track ship tracks, with significant application effects.
2023,45(15): 131-134 收稿日期：2023-02-27
DOI：10.3404/j.issn.1672-7649.2023.15.026
分类号：TP391
基金项目：江西省高等学校教学改革研究课题(JXJG-22-24-4)
作者简介：周延木(1978-),男,硕士,讲师,从事计算机三维设计及视觉传达技术等研究
参考文献：
[1] 雷禹, 冷祥光, 计科峰. 基于Google Earth Engine的海量舰船目标SAR图像处理应用研究[J]. 信号处理, 2021, 37(6): 1075–1085
[2] 刘万军, 高健康, 曲海成, 等. 多尺度特征增强的遥感图像舰船目标检测[J]. 自然资源遥感, 2021, 33(3): 97–106
[3] 向涛. 基于显著区域提取的红外图像舰船目标检测[J]. 电讯技术, 2020, 60(7): 785–790
[4] 王春静, 刘丽, 谭艳艳, 等. 基于模糊颜色特征和模糊相似度的图像检索方法[J]. 计算机科学, 2021, 48(8): 191-199.
[5] 刘凯, 汪兴海, 张杰. 基于深度玻尔兹曼机的图像多特征融合[J]. 舰船电子工程, 2020, 40(1): 32–36
[6] 马健, 史文旭, 鲍胜利. 基于特征融合SSD的遥感图像舰船目标检测[J]. 计算机应用, 2019, 39(z2): 253–256
[7] 刘洁瑜, 赵彤, 刘敏. 基于RetinaNet的SAR图像舰船目标检测[J]. 湖南大学学报(自然科学版), 2020, 47(2): 85–91
[8] 严春满, 王铖. 基于特征增强的SAR图像舰船小目标检测算法[J]. 控制与决策, 2023, 38(1): 239-247.
