二阶段沙尘降质图像增强算法

Two-stage dust-degraded image enhancement algorithm

摘要

针对沙尘降质图像存在色偏、对比度低和能见度差等问题，提出二阶段沙尘降质图像增强算法。算法包含沙尘图像颜色校正算法和基于残差融合的尘雾去除网络。第1阶段提出在沙尘降质图像的Lab颜色空间中用以图像饱和度为权重的加权灰色世界理论进行颜色校正，有效解决沙尘降质图像色偏问题。第2阶段设计基于残差融合的尘雾去除网络提升图像对比度和清晰度。实验结果表明，算法可以有效去除色偏问题，并在提高图像对比度的同时增强图像细节的可见度，相比对比实验结果，该文算法的PSNR和SSIM分别提高2.338 0%和3.066 2%。

Abstract

To address the issues of color bias, low contrast and poor visibility in dust-degraded images, a two-stage dust-degraded image enhancement algorithm is proposed. The algorithm comprises a dust-degraded image color correction algorithm and a residual fusion-based haze removal network. In the first stage, a weighted gray-world theory based on image saturation in the Lab color space is proposed for color correction, effectively addressing the color bias issue in dust-degraded image. In the second stage, a residual fusion-based dust and haze removal network is designed to enhance the contrast and clarity of the images. The experimental results show that the algorithm can effectively remove color bias and enhance the visibility of image details while improving image contrast. Compared to the best results from the comparative experiments, the proposed algorithm improves PSNR and SSIM by 2.338 0% and 3.066 2%, respectively.

关键词

沙尘降质图像 / 图像增强 / 加权灰色世界理论 / 颜色校正 / 残差融合

Key words

dust-degraded image / image enhancement / weighted gray-world theory / color correction / residual fusion

参考文献

[1] SHI Z, FENG Y, ZHAO M, et al. Normalised gamma transformation based contrast limited adaptive histogram equalisation with colour correction for sand—dust image enhancement[J]. IET Image Processing, 2020, 14(4): 747-756.

[2] PARK T H, EOM I K. Sand—dust image enhancement using successive color balance with coincident chromatic histogram[J]. IEEE Access, 2021, 9: 19749-19760.

[3] ZOTIN A G. Fast algorithm of image enhancement based on multi—scale retinex[J]. International Journal of Reasoning—Based Intelligent Systems, 2020, 12(2): 106-116.

[4] NARASIMHAN S G, NAYAR S K. Vision and the atmosphere[J]. International Journal of Computer Vision, 2002, 48(3): 233-254.

[5] HE K, SUN J, TANG X. Single image haze removal using dark channel prior[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 33(12): 2341-2353.

[6] DHARA S K, ROY M, SEN D, et al. Color cast dependent image dehazing via adaptive airlight refinement and nonlinear color balancing[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2020, 31(5): 2076-2081.

[7] GAO G X, LAI H C, JIA Z H, et al. Sand—dust image restoration based on reversing the blue channel prior[J]. IEEE Photonics Journal, 2020, 12(2): 1-16.

[8] SHI Z, FENG Y, ZGAO M, et al. Let you see in sand dust weather: A method based on halo—reduced dark channel prior dehazing for sand—dust image enhancement[J]. IEEE Access, 2019, 7: 116722-116733.

[9] SI Y, YANG F, LIU Z. Sand dust image visibility enhancement algorithm via fusion strategy[J]. Scientific Reports, 2022, 12(1): 13226.

[10] GAO Y, XU W, LU Y. Let you see in haze and sandstorm: Two—in—one low visibility enhancement network[J]. IEEE Transactions on Instrumentation and Measurement, 2023, 72: 1-12.

[11] 丁元, 邬开俊 . 多分支修复网络的沙尘降质图像增强算法[J]. 计算机工程与应用, 2023, 59(24): 227-237.

[12] DING Y, WU K J. Sand dust degradation images enhancement algorithm via multi—branch restoration network[J]. Computer Engineering and Applications, 2023, 59(24): 227-237.

[13] CUI Y, KNOLL A. Exploring the potential of channel interactions for image restoration[J]. Knowledge—Based Systems, 2023, 282: 111156.

[14] QIN X, WANG Z, BAI Y, et al. FFA—Net: Feature fusion attention network for single image dehazing[C]// Proceedings of the AAAI Conference on Artificial Intelligence. New York: AAAI, 2020: 11908-11915.

[15] WU H, QU Y, LIN S, et al. Contrastive learning for compact single image dehazing[C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. [S. l.]: IEEE, 2021: 10551-10560.

[16] ZHENG Y, ZHAN J, HE S, et al. Curricular contrastive regularization for physics—aware single image dehazing[C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Vancouver: IEEE, 2023: 5785-5794.

[17] BUCHSBAUM G. A spatial processor model for object colour perception[J]. Journal of the Franklin Institute, 1980, 310(1): 1-26.

[18] LIM B, SON S, KIM H, et al. Enhanced deep residual networks for single image super—resolution[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. [S. l.]: IEEE, 2017: 136-144.

[19] WANG X, HAN X, HUANG W, et al. Multi—similarity loss with general pair weighting for deep metric learning[C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach: IEEE, 2019: 5022-5030.

[20] SAHU S, SINGH A K, GHRERA S P, et al. An approach for denoising and contrast enhancement of retinal fundus image using CLAHE[J]. Optics & Laser Technology, 2019, 110: 87-98.

基金资助

内蒙古自治区重点研发与成果转化计划(2023YFSH0043)

内蒙古自治区重点研发与成果转化计划(2023YFDZ0043)

全国高等院校计算机基础教育研究会项目(2023-AFCEC-039)

甘肃省教育厅高校教师创新基金项目(2025A-054)