Enhancing Underwater Images through Deep Curve Estimation

깊은 곡선 추정을 이용한 수중 영상 개선

  • Muhammad Tariq Mahmood (Korea University of Technology and Education, School of Computer Science and Engineering) ;
  • Young Kyu Choi (Korea University of Technology and Education, School of Computer Science and Engineering)
  • Received : 2024.05.01
  • Accepted : 2024.06.21
  • Published : 2024.06.30

Abstract

Underwater images are typically degraded due to color distortion, light absorption, scattering, and noise from artificial light sources. Restoration of these images is an essential task in many underwater applications. In this paper, we propose a two-phase deep learning-based method, Underwater Deep Curve Estimation (UWDCE), designed to effectively enhance the quality of underwater images. The first phase involves a white balancing and color correction technique to compensate for color imbalances. The second phase introduces a novel deep learning model, UWDCE, to learn the mapping between the color-corrected image and its best-fitting curve parameter maps. The model operates iteratively, applying light-enhancement curves to achieve better contrast and maintain pixel values within a normalized range. The results demonstrate the effectiveness of our method, producing higher-quality images compared to state-of-the-art methods.

Keywords

Acknowledgement

이 논문은 2023년도 한국기술교육대학교 교수 교육연구진흥과제 지원에 의하여 연구되었음.

References

  1. M. J. Islam, Y. Xia and J. Sattar, "Fast Underwater Image Enhancement for Improved Visual Perception," IEEE Robotics and Automation Letters, vol. 5, no. 2, pp. 3227-3234, 2020. https://doi.org/10.1109/LRA.2020.2974710
  2. He, K., Sun, J., Tang, X. Single Image Haze Removal Using Dark Channel Prior. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 33(12), 2341-2353. https://doi.org/10.1109/TPAMI.2010.168
  3. P. Drews, E. Nascimento, F. Moraes, S. Botelho and M. Campos, "Transmission Estimation in Underwater Single Images," Paper presented at the Proceedings of the IEEE International Conference on Computer Vision Workshops, 2013.
  4. C. O. Ancuti, C. Ancuti, C. De Vleeschouwer and P. Bekaert, "Color Balance and Fusion for Underwater Image Enhancement," IEEE Transactions on Image Processing, vo. 27, no. 1, pp. 379-393, 2017. https://doi.org/10.1109/TIP.2017.2759252
  5. Berman, D., Treibitz, T., Avidan, S. Single Image Dehazing Using Haze-Lines. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 42(3), 720-734. https://doi.org/10.1109/TPAMI.2018.2882478
  6. Li, C. Y., Guo, J. C., Cong, R. M., Pang, Y. W., Wang, B. Underwater Image Enhancement by Dehazing with Minimum Information Loss and Histogram Distribution Prior. IEEE Transactions on Image Processing, 2016, 25(12), 5664-5677. https://doi.org/10.1109/TIP.2016.2612882
  7. Peng, Y. T., Cosman, P. C. Underwater Image Restoration Based on Image Blurriness and Light Absorption. IEEE Transactions on Image Processing, 2017, 26(4), 1579-1594. https://doi.org/10.1109/TIP.2017.2663846
  8. Li, B., Peng, X., Wang, Z., Xu, J., Feng, D. Aod-Net: All-in-One Dehazing Network. Paper presented at the Proceedings of the IEEE International Conference on Computer Vision, 2017.
  9. Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., Bengio, Y. Generative Adversarial Nets. Advances in Neural Information Processing Systems 27, 2014.
  10. Yang, M., Hu, K., Du, Y., Wei, Z., Sheng, Z., Hu, J. Underwater Image Enhancement Based on Conditional Generative Adversarial Network. Signal Processing: Image Communication, 2020, 81, 115723.
  11. Yang, H. H., Fu, Y. Wavelet U-Net and the Chromatic Adaptation Transform for Single Image Dehazing. Paper presented at the 2019 IEEE International Conference on Image Processing (ICIP), 2019.
  12. M. T. Mahmood and Y. K. Choi, "Mutual Structures based Improvement in Depth from Focus," Journal of the Semiconductor & Display Technonogy, vol. 21, no. 3, pp. 17-21, 2022.