Research on ship identification methods in optical remote sensing images
There are many interferences in the detection and recognition process of ship targets in optical remote sens-ing images,which make it difficult to accurately identify ship targets.Therefore,research is conducted on ship recognition methods in optical remote sensing images.Firstly,extract the salient regions of ship targets in optical remote sensing images,suppress the influence of background information such as clouds,sea clutter,and sea land on ship target recognition,and complete rough recognition of ship targets in optical remote sensing images.Then,based on the extracted salient regions of optical remote sensing images,a CNN network is used for precise recognition of ship targets.The experimental results show that the design method can effectively extract the salient regions of ship targets in optical remote sensing images,and extract the ship target features in the salient regions;The accuracy of ship target recognition is always above 95％,which is practical.
NETL
NSTL
万方数据
