为了保证船舶的安全运行，资源的合理利用，避免海上交通安全事故的发生，提出基于遥感图像的船舶交通流量预测方法。结合主成分分析法、多层元胞更新方法等建立船舶遥感图像的显著模型，获取船舶遥感图像的显著图；通过恒虚警(CFAR)检测方法在船舶遥感图像的显著图上进行遍历，完成船舶的目标检测；依据船舶目标检测结果，统计船舶数量，计算海上船舶的拥堵率，并建立支持向量机(SVM)模型，将船舶的拥堵率作为SVM模型的输入，与输出数据的结果进行映射，实现船舶交通流量的预测。实验结果表明：该方法可准确地预测出船舶交通流量；该方法不受海杂波等干扰的影响，可准确地检测出船舶目标；在云雾和海杂波的环境干扰下，该方法预测船舶交通流量结果的均等系数趋近于0，预测能力较强。
In order to ensure the safe operation of ships, rational use of resources and avoid the occurrence of maritime traffic accidents, a ship traffic flow prediction method based on remote sensing images is proposed. The salient model of ship remote sensing image is established by combining principal component analysis and multi-layer cell updating method, and the salient map of ship remote sensing image is obtained. Constant false alarm (CFAR) detection method is used to traverse the significant map of the ship remote sensing image to complete the target detection of the ship. According to the ship target detection results, the number of ships is counted, the congestion rate of ships at sea is calculated, and the support vector machine (SVM) model is established. The congestion rate of ships is used as the input of the SVM model, and the results of the output data are mapped to realize the prediction of ship traffic flow. The experimental results show that this method can accurately predict the ship traffic flow. The method is not affected by sea clutter and can accurately detect the ship target. Under the environment interference of cloud, fog and sea clutter, the uniformity coefficient of the forecast results of ship traffic flow is close to 0, and the prediction ability is strong.
2023,45(17): 162-165 收稿日期：2023-05-16
DOI：10.3404/j.issn.1672-7649.2023.17.032
分类号：TP391
基金项目：江苏航运职业技术学院科研课题（HYKY/2020B03）
作者简介：曾晓晴(1983-),女,硕士,讲师,研究方向为物流管理及交通运输管理
参考文献：
[1] 李佩钰. 一种基于小波和神经网络的短时交通流量预测[J]. 计算机技术与发展, 2020, 30(1): 135-139.
[2] 黄振盛, 汪玉美, 韩江洪, 等. 基于MLS-SVM和时空特性的短时交通流量预测方法[J]. 合肥工业大学学报(自然科学版), 2020, 43(1): 57-63.
[3] 薛晗, 邵哲平, 潘家财, 等. 基于文化萤火虫算法-广义回归神经网络的船舶交通流量预测[J]. 上海交通大学学报, 2020, 54(4): 421-429.
[4] 闫少华, 谢晓璇, 张兆宁. 基于小波优化GRU-ARMA模型的空中交通流量短时预测方法[J]. 交通信息与安全, 2022, 40(4): 177-184.
[5] 祁朵, 毛政元. 基于自适应时序剖分与KNN的短时交通流量预测[J]. 地球信息科学学报, 2022, 24(2): 339-351.
[6] 郑义彬, 赖伟伟. 基于支持向量机的高速短时交通流量预测[J]. 工程与建设, 2020, 34(2): 201-204.
