|
|
Research on Extraction Method of Coastal Aquaculture Areas on High Resolution Remote Sensing Image based on Multi-features Fusion |
Cheng Bo1,4,Liu Yueming1,3,Liu Xunan2,Wang Guizhou1,Ma Xiaoxiao1,3 |
(1.Institute of Remote Sensing and Digital Earth Chinese Academy of Sciences,Beijing 100094,China;
2.National Marine Hazard Mitigation Service,Beijing 100194,China;
3.University of Chinese Academy of Sciences,Beijing 100049,China;
4.Sanya Institute of Remote Sensing,Sanya 572029,China) |
|
|
Abstract Coastal aquaculture areas are the important marine disaster bearing body,it is of great significance to carry out the research about automatically extraction aquaculture areas based on remote sensing for master basic information of coastal areas,mitigation and prevention of marine disasters.Used Gaofen-2 images for experimental data,and chose Dongshan Island sea in Fujian province as the experimental area,on the basis of analyzed the spectral characteristics of the aquaculture areas,construct feature index to extract spectral feature,used gray-level co-occurrence matrix method to extract the texture feature of the aquaculture areas,fused spectral features and texture features after feature selection,then used Otsu method to determine the threshold for raft culture area and fishing cage culture area,to achieve high precision intelligent extraction and classification for coastal aquaculture areas.The extraction accuracy of the raft culture area is more than 80%,and the extraction precision of the fish cage culture area is above 90%,and the overall extraction precision of the culture area is up to 87%.
|
Received: 03 March 2017
Published: 15 May 2018
|
|
No Suggested Reading articles found! |
|
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
|
Shared |
|
|
|
|
|
Discussed |
|
|
|
|