Underwater image restoration and enhancement using depth estimation and gradient descent
Due to inherent scattering and absorption,underwa-ter images inevitably suffer from multiple degradations arising from blurring,low contrast and color distortion,thereby seri-ously deteriorating visual perception. In this paper,a deep learning-based underwater image restoration and enhancement framework (UIRENet) was proposed by virtue of depth esti-mation and gradient descent strategy. With the aid of convolu-tional and nonlinear activation function modules,a deep per-ception network was constructed to achieve scene depth per-ception maps for different degradation regions,thereby overco-ming the dependence of scene-depth degradation. A gradient optimization strategy was further proposed to optimize the pa-rameters of convolutional networks and improve the perform-ance of deep network enhancement. Combined with perceptu-al,edge and underwater color constancy losses,a comprehen-sive loss function for underwater image enhancement networks was rationally formed. Comprehensive experiments on the UIEB-90,UIEB-M and EUVP datasets show that the UI-RENet framework significantly outperforms typical underwater image enhancement methods in terms of reducing underwater image blurriness and improving visual effects. In particular,comparing to CLAHE,ICM,GC,IBLA,DCP,ULAP,FUnIE-GAN,UGAN and Uformer,the objective evaluation metric UIQM can be promoted by 0.3700,0.6446,0.5919,1.3081,1.3032,1.1672,0.0593,0.1329 and 0.0934,re-spectively.
万方数据
维普
