| 基于深度学习RCF模型的三都澳筏式养殖区提取研究 |
| |
| 引用本文： | 刘岳明,杨晓梅,王志华,陆尘.基于深度学习RCF模型的三都澳筏式养殖区提取研究[J].海洋学报,2019,41(4):119-130. |
| |
| 作者姓名： | 刘岳明 杨晓梅 王志华 陆尘 |
| |
| 作者单位： | 中国科学院地理科学与资源研究所 资源与环境信息系统国家重点实验室,北京 100101;中国科学院大学,北京100049;中国科学院地理科学与资源研究所 资源与环境信息系统国家重点实验室,北京,100101 |
| |
| 基金项目： | 中国科学院战略性先导科技专项（A类）（XDA19060303）；国家重点研发计划项目（2016YFC1402003）；国家自然科学基金（41671436）；资源与环境信息系统国家重点实验室自主创新项目（O88RAA01YA）。 |
| |
| 摘 要： | 三都澳是中国重要的海水养殖海湾，在水产养殖中占有较高的经济地位。快速准确地获取养殖区的分布范围、数量以及面积等信息，对养殖区规划、产值估计、生态调查、风暴潮灾害预防等具有重要的意义。然而，随着养殖区域的扩大，海水背景状态愈发复杂多样，光谱特征差异较大，为养殖区提取带来困难。在本实验中，利用高分辨率遥感卫星GF-2图像，引入深度学习RCF（Richer Convolutional Features）网络模型对海湾内的筏式养殖区进行了提取。结果显示:该方法无需事先对区域进行水陆分离处理，且对水中泥沙较多的区域以及海浪较大的区域有很好的提取效果，提取精度达93%以上，适合进行大规模海水养殖区提取应用。
|
| 关 键 词： | 筏式养殖区 RCF模型 深度学习 高分辨率遥感 |
| 收稿时间： | 2018/4/18 0:00:00 |
| 修稿时间： | 2018/9/11 0:00:00 |
Extracting raft aquaculture areas in Sanduao from high-resolution remote sensing images using RCF |
| |
| Liu Yueming,Yang Xiaomei,Wang Zhihua and Lu Chen.Extracting raft aquaculture areas in Sanduao from high-resolution remote sensing images using RCF[J].Acta Oceanologica Sinica (in Chinese),2019,41(4):119-130. |
| |
| Authors: | Liu Yueming Yang Xiaomei Wang Zhihua and Lu Chen |
| |
| Institution: | 1.State Key Laboratory of Resources and Enviromental Information System, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China;University of Chinese Academy of Sciences, Beijing 100049, China2.State Key Laboratory of Resources and Enviromental Information System, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China |
| |
| Abstract: | Sanduao is an important sea-breeding bay in China and holds a high economic status in aquaculture. Quickly and accurately obtaining information such as the distribution area, quantity, and aquaculture area is important for breeding area planning, production value estimation, ecological survey, and storm surge prevention. However, as the aquaculture area expands, the seawater background becomes increasingly complex and spectral characteristics differ dramatically, making it difficult to determine the aquaculture area. In this study, we used a high-resolution remote-sensing satellite GF-2 image to introduce a deep-learning Richer Convolutional Features network model to extract the raft aquaculture area in the bay. The results demonstrate that this method does not require land and water separation of the area in advance, and good extraction can be achieved in areas with more sediment and waves, with an extraction accuracy >93%, which is suitable for large-scale raft aquaculture area extraction. |
| |
| Keywords: | raft aquaculture area RCF model deep learning high-resolution remote sensing |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《海洋学报》浏览原始摘要信息 |
| 点击此处可从《海洋学报》下载免费的PDF全文 |
|
