Research on enhancement of foggy ship image based on convolutional neural network
Ship image enhancement plays an important role in feature extraction of target image.In this paper,the im-age model of foggy ship is studied,and the algorithm flow of foggy ship image enhancement based on convolutional neural network is designed,which includes four stages:image data preparation,convolutional neural network model construction,model training and image enhancement.Finally,the effects before and after ship image enhancement are compared,and the background and details of ship image in foggy day are enhanced.The results show that the convolutional neural network based image enhancement method proposed in this paper is effective and can effectively improve the image quality.
convolutional neural networkfoggyimage enhancementloss function
NETL
NSTL
万方数据
