基于无人机紫外与SAR的溢油遥感监测方法研究

Marine oil pollution will not only cause huge economic losses, but also bring damages to the ecological environment which is difficult to be repaired. Accurate and efficient oil spill monitoring is still a serious problem need to be solved currently..Ultraviolet (UV) sensors which are very sensitive to the oil film can quickly find the oil film, but there will be misjudgment; and the accuracy of oil spill remote sensing detection using SAR is very high. So the combination of these can detect oil spill accurately. Unmanned Aerial Vehicle (UAV) platform can achieve fast emergency response of oil spill inexpensively. The load weight of UAV SAR and UV sensor is light, and they can both be integrated into the UAV to the joint exploration of oil spill for meeting the requirement of operational monitoring, but the research about this aspect has not been any related reports yet..The objective of this project will be designed to study the features of UV images and the shapes, textures, polarization scattering characteristics of SAR images in many different situations such as different types and thickness of oil spill and marine environmental conditions, to construct a feature database of oil spill, and to establish a joint oil spill detecting method of SAR and UV sensor based on feature combination; Then the transformation program of acquisition mode and control unit will be considered, and the method of efficiently acquire UAV SAR and UV images of oil spill will be realized. Situ measurement and remote sensing observation using UAV SAR and UV sensor will be designed in the sea near the pier of National Deep Sea Center, for establishing the oil spill feature database and verifying the effectiveness of oil spill detecting method.

溢油污染不仅会造成巨大的经济损失，而且给生态环境带来难以修复的破坏。准确、高效地监测海面溢油仍是当前亟需解决的问题。紫外传感器对油膜非常敏感，可快速发现，但存在误判；而SAR溢油探测的精度较高，两者相结合可准确探测溢油。无人机平台可低成本地实现溢油快速应急响应，无人机载SAR和紫外传感器的载荷重量小，可同时集成于无人机上开展联合溢油探测，以满足业务化监测需求，此方面的研究尚未见有相关报道。.本项目拟研究溢油不同种类、厚度、在不同海洋环境条件下的紫外图像特征和SAR纹理特征、形状特征、散射特征，构建溢油特征数据库，并建立一种基于特征组合的溢油SAR与紫外联合探测方法；在此基础上研究对无人机数据获取模式和控制单元等的改造方案，进而实现溢油SAR和紫外图像的高效获取。拟在国家深海基地码头附近海域开展溢油现场测量和无人机SAR与紫外遥感联合探测实验来建立溢油特征数据库并验证溢油探测算法的有效性。

溢油污染不仅会造成巨大的经济损失，而且给生态环境带来难以修复的破坏。准确、高效地监测海面溢油仍是当前亟需解决的问题。紫外传感器对油膜非常敏感，可快速发现，但存在误判；而SAR溢油探测的精度较高，两者相结合可准确探测溢油。. 本项目在2019年和2020年分别进行了两次溢油多手段联合探测实验，建立了不同种类、厚度条件下溢油的紫外图像特征库和SAR纹理特征、形状特征、散射特征，构建了溢油SAR特征数据库，发展了基于多特征多核学习的全极化SAR溢油检测方法、基于DRLSE模型的SAR溢油检测方法、基于GA-WNN的极化SAR溢油检测方法和基于光谱信息与纹理特征结合的光学溢油检测方法等。实现了溢油的SAR与光学的多手段快速、准确检测。.