Underwater image processing has attracted increasing attention, however the algorithms are often evaluated using different test images of different resolutions, making comparison of results difficult. Also, the results are often evaluated subjectively.


This website is dedicated to underwater image quality evaluation. This is an easy-to-use platform for researchers to evaluate and compare their work by providing the following -

  1. collection of standard underwater image datasets, and
  2. online underwater image quality evaluation using UIQM [1] and UCIQE [2], with code provided by the authors

Getting started

Simply log in and start uploading images to public leaderboard or as private submission!


Choice of evalution metric

We have chosen 2 metrics developed for underwater image quality evaluation [1][2], however it has been discussed that these measures cannot account for all distortions [3], thus suggesting there is still room for improvement. We hope this public platform will help us gain a better understanding of underwater image quality evaluation.


New metric?

Please contact us if you developed a new metric or think any metric should be added to the platform.


Publication

This website is accompanied by the paper 'An online platform for underwater image quality evaluation', published in Workshop on Computer Vision for Analysis of Underwater Imagery (CVAUI), Beijing, China, August 2018. The paper discusses the common problems faced in comparing underwater image processing methods, and details the metrics used and datasets provided by the platform.


References

  1. K. Panetta, C. Gao and S. Agaian, "Human-visual-system-inspired underwater image quality measures," IEEE J. Ocean. Eng., vol. 41 no. 3 pp. 541-551 Jul. 2015.
  2. M. Yang and A. Sowmya, β€œAn underwater color image quality evaluation metric,” IEEE Trans. Image Process., vol. 24, no. 12, pp. 6062–6071, Dec. 2015.
  3. S. Emberton, L. Chittka, and A. Cavallaro, "Underwater image and video dehazing with pure haze region segmentation," Computer Vision and Image Understanding, ISSN 1077-3142 2017.