Due to the absorption and scattering of light in water, underwater images commonly exhibit degradation phenomena such as color cast, low visibility, and blurred details. In response to these issues, we propose a color, detail, contrast, and multi-scale fusion underwater image enhancement algorithm called CDCM. The algorithm first uses a color restoration method based on dark and bright channels to effectively correct color distortion of underwater images and restore their natural color balance. Secondly, utilizing morphological operations to enhance the contour and structural information of objects in the image so as to improve detail representation. In addition, the black eagle optimizer (BEO) is introduced and a new fitness function is designed to adaptively optimize image contrast. In the fusion stage, principal component weights are proposed and combined with other weighting strategies to achieve multi-scale image information fusion, enhancing the contrast while preserving rich textures and details. Experimental results on two real underwater image datasets UIEB and RUIE demonstrate that our method effectively reduces degradation phenomena, with image enhancement by improvements in color fidelity, contrast, and detail clarity compared to the existing methods. In terms of objective indicators, our method is also superior to other relevant methods, such as UCIQE, UIQM, AG, IE, PCQI, etc. Our work contributes to advancing underwater image processing techniques.

Aiming at the problem of ringing artifacts existing in the edge of image in traditional blind image deblurring methods, l₁/l₂ regularization-based blind image deblurring method is proposed. The latent image is constrained by l₁/l₂ regularization, and the two-norm constraint is applied to the blur kernel to remove the noise of the blur kernel. During the solution process, the latent image and the blur kernel are updated alternately anditeratively, and the deblur redimage is finally obtained by combining the finest estimated blur kernel with the non-blind deblurring method. The experimental results show that the proposed method improves the quality of image deblurring and effectively removes some ringing artifacts. It has a good restoration effect on natural blurred images.

Photographs taken in daily life often became blurred due to shaking, out-of-focus, changes in depth of field, and movement of photographed objects. Aiming at this problem, a double-channel cyclic image deblurring method based on edge features was proposed. Firstly, image edge gradient operator was introduced as a threshold based on the rule that the maximum value of the image edge gradient will decrease after the blurring process, making the blurred image be divided into two channels: edge channel and non-edge channel. Secondly, a double-channel loop iteration network was designed, where the edge gradient was used in the edge channel to sample the main edge structure and bilateral filtering was used in the non-edge channel to extract the detailed texture feature information. Finally, the feature information extracted from two channels was cyclically iterated to obtain a clear image using the deblurring model with maximum a posteriori probability. The experimental results showed that the image evaluation indexes obtained by the proposed deblurring model were superior to those of other algorithms, and the edge structure and texture details of the image were effectively recovered with better performance.

Corona Virus Disease 2019(COVID-19) is one of the global concerns due to its highly infectious and highly pathogenic coronavirus. It is of great value to effectively predict the cumulative number of confirmed cases of COVID-19 for the prevention and control of COVID-19. In this paper, the weighted average salp swarm algorithm is proposed, named by AVSSA, whose validation is performed by 23 benchmark functions. Then AVSSA is utilized to optimize the parameters of BP neural network to establish the predicted model AVSSA-BP for predicting the COVID-19. The experimental results show that the predicted model AVSSA-BP has the least errors and the highest coefficient of determination. Therefore, the proposed AVSSA is an effective algorithm.