Xu, L. and Xu, X., 2019. Regular super-resolution restoration algorithm based on fuzzy similarity fusion for a single haze image over the ocean. In: Guido-Aldana, P.A. and Mulahasan, S. (eds.), Advances in Water Resources and Exploration. Journal of Coastal Research, Special Issue No. 93, pp. 489–494. Coconut Creek (Florida), ISSN 0749-0208.

The current image restoration algorithms mainly use a single method to restore haze images over the ocean. It is impossible to perform super-resolution restoration, and at the same time, it is impossible to aggregate the advantages of various algorithms, which has some limitations. As a result, the image restoration performance is not good, and the effect is poor. To address this problem, a regular super resolution restoration algorithm for single fog haze images over the sea based on fuzzy similarity fusion is proposed in this paper. The non-local mean of the discrete cosine transform is used to denoise the haze images over the sea and keep the smooth part of the smog image and most of the details. The fuzzy similarity fusion algorithm is used to fuse the haze images over the sea after denoising, and the pixels of the restored image with a certain similarity is weighted and averaged. The weight is determined by the regional feature of the pixel, and finally the fusion image is obtained. Combining the haze images over the sea and the regular super resolution restoration algorithm, the priori information of the sequence image is used to effectively suppress the ringing effect of the restored image, and achieve the regular super resolution restoration of the single haze over a single sea. Experimental results show that the proposed algorithm can effectively improve the restoration effect of haze images over the sea.