Please contact us if you would prefer your dataset to be linked from this page.


ReefnFish

Im1.pngShip.pngfish.pngreef1.pngreef2.pngreef3.png
click to download dataset as .zip
Number of images in the dataset: 6
Description: Images used in different papers such as [1] [2].

Bali

diver.pngfish1.pngfish2.pngfish3.png
click to download dataset as .zip
Number of images in the dataset: 4
Description: Description: This dataset consists of 4 images extracted from Diving in Bali (720p), appeared in [3]. Agreed to use by BubbleVision.

CVAUI_test

CIS.pngHE.pngRS.pnginput.png
click to download dataset as .zip
Number of images in the dataset: 4
Description: This is a test dataset created for CVAUI 2018 and consists of 4 images: input.png - input; HE.png - histogram equalisation; RS.png - range stretching; CIS.png - processed by CIS QMUL.

More datasets will be added soon. Stay tuned.


References

  1. C. O. Ancuti, C. Ancuti, C. De Vleeschouwer, and P. Bekaert, “Color balance and fusion for underwater image enhancement,” in IEEE Trans. Image Process., 2018, vol. 27, pp. 379–393.
  2. A. Galdran, D. Pardo, A. Picon, and A. Alvarez-Gila, “Automatic red-channel underwater image restoration,” in J. Vis. Commun. Image Represent., 2015, vol. 26, pp. 132–145
  3. J. Chiang, and Y. Chen, “Underwater image enhancement by wavelength compensation and dehazing,” in IEEE Trans. Image Process., 2012, vol. 21, pp. 1756–1769