A referenceless image degradation perception method based on the underwater imaging model

摘要

With the development of underwater image processing, more and more underwater image restoration algorithms have been proposed. To date, most underwater image degradation perception methods are ineffective in multi-color environments. In this paper, we proposed a referenceless image degradation perception method based on the underwater imaging model. Our method includes the colorfulness index, contrast index, and sharpness index. The colorfulness index is used to measure the color loss. The contrast index and sharpness index are used to measure the blurring. For the contrast index and sharpness index, we proposed a grayscale conversion method that can adaptively adjust the coefficients of red, green, and blue (RGB) values. Their generalization ability under multi-colored environments can be enhanced by the proposed adaptive grayscale conversion method. It is useful for underwater images that are normally dominated by green and blue color channels. Compared with the leading evaluation metrics available in the literature, our proposed method can better perceive the degradation of underwater images. No matter under what lighting conditions, our proposed three indexes can decrease with the degradation of images. It makes our proposed three indexes can be effective in both general underwater scenarios and hash scenarios. The correlation coefficients of our proposed indexes are the largest. More importantly, the proposed indexes can be used to process underwater images in real time and evaluate the performance of underwater image restoration algorithms. Our proposed indexes can also be applied to monitor the turbidity of water and improve the underwater image restoration algorithms. A single proposed index cannot comprehensively perceive the quality of underwater images. The proposed three indexes need to be integrated into an evaluation metric in future work.