Double light curtain-constrained hazy image restoration algorithm based on improved atmospheric scattering model
-
摘要:
当雾霾退化场景处于光照不均匀的条件下时，部分场景细节不仅会由于雾气遮盖导致可见度降低，同时会因为光照阴影使得部分区域不可见。针对这一问题，提出一种基于改进型大气散射模型的双光幕边界约束的雾天图像复原算法。分析传统大气散射模型的成像原理，利用其退化机理结合Retinex理论对模型进行改进；引入均值不等关系与高斯衰减函数，通过预估特征值的方法对大气光幕进行估计，并设定上下边界对其进行约束；依照改进型大气散射模型求取场景入射光，并利用亮通道先验求取有雾图像亮通道对场景入射光进行补偿；改进局部大气光的获取方法，提出基于中通道的局部大气光估计方法，结合所求大气光幕与场景入射光代入改进型大气散射模型获得无雾图像，并使其与图像纹理层进行融合得到最终的复原结果。根据对实验结果的定性与定量分析，所提算法不仅可以有效复原出场景光照不均的有雾图像，针对其余场景下的雾霾场景也可得到较好的复原效果，且复原场景清晰，明亮度适宜。
Abstract:When a haze-degraded scene is under uneven lighting conditions, some scene details will be less visible due to hazy obscuration and light shadows. To address this issue, a hazy image restoration algorithm with double light curtain boundary constraints based on an improved atmospheric scattering model was proposed. The imaging principle of the conventional atmospheric scattering model was analyzed, and the model was improved by using its degradation mechanism combined with the Retinex theory. The mean inequality relation and Gaussian decay function were introduced to estimate the eigenvalue of the atmospheric light curtain. In addition, the upper and lower boundaries were set to constrain the estimated values. The scene-incident light was obtained by an improved atmospheric scattering model, which was compensated by using the bright channel of the hazy image through the bright channel prior method. The method of local atmospheric light acquisition was improved. A mid-channel-based local atmospheric light estimation method was proposed. The estimated atmospheric light curtain and scene-incident light were introduced into an improved atmospheric scattering model to obtain a hazy-free image. Then, the hazy-free image was fused with the image texture layer to obtain the final restored image. According to the subjective and objective analyses of the experimental results, the proposed algorithm can not only effectively restore the hazy images with uneven scene illumination but also get a better restoration effect for the haze scenes. The restored scene is clear, and the brightness is moderate.
-
图 2 大气散射模型物理退化过程
Figure 2. Physical degradation process of atmospheric scattering model
图 3 2种大气散射模型的复原效果对比
Figure 3. Comparison of restoration effects of two atmospheric scattering models
图 5 大气光幕估计和复原结果示意图
Figure 5. Schematic diagram of atmospheric light curtain estimation and restoration results
图 7 大气光图及其复原效果对比
Figure 7. Comparison of atmospheric light maps and their restoration effects
图 8 纹理层图像及添加前后复原效果对比
Figure 8. Comparison of texture layer images and restoration effects before and after adding
图 10 不同算法下光照不均匀图像的复原结果
Figure 10. Restoration results of images with uneven light under different algorithms
图 11 不同算法下颜色变化剧烈图像的复原结果
Figure 11. Restoration results of images with drastic color changes under different algorithms
图 12 不同算法下远景含天空图像的复原结果
Figure 12. Restoration results of distant view with sky images under different algorithms
图 13 不同算法下近景含人物图像的复原结果
Figure 13. Restoration results of close shot images with people under different algorithms
图 14 不同算法下RESIDE测试集图像的复原结果
Figure 14. Restoration results of RESIDE test set images under different algorithms
图 15 各个算法的定量分析指标对比
Figure 15. Comparison of quantitative analysis indexes of each algorithms
-
[1] 蒋华伟, 杨震, 张鑫, 等. 图像去雾算法研究进展[J]. 吉林大学学报(工学版), 2021, 51(4): 1169-1181.JIANG H W, YANG Z, ZHANG X, et al. Research progress of image dehazing algorithms[J]. Journal of Jilin University (Engineering and Technology Edition), 2021, 51(4): 1169-1181 (in Chinese). [2] 江巨浪, 刘国明, 朱柱, 等. 基于快速模糊聚类的动态多直方图均衡化算法[J]. 电子学报, 2022, 50(1): 167-176. doi: 10.12263/DZXB.20201040JIANG J L, LIU G M, ZHU Z, et al. Dynamic multi-histogram equalization based on fast fuzzy clustering[J]. Acta Electronica Sinica, 2022, 50(1): 167-176 (in Chinese). doi: 10.12263/DZXB.20201040 [3] XIAO L M, LI C, WU Z Z, et al. An enhancement method for X-ray image via fuzzy noise removal and homomorphic filtering[J]. Neurocomputing, 2016, 195: 56-64. doi: 10.1016/j.neucom.2015.08.113 [4] LIU H T, LU H Q, ZHANG Y. Image enhancement for outdoor long-range surveillance using IQ-learning multiscale Retinex[J]. IET Image Processing, 2017, 11(9): 786-795. doi: 10.1049/iet-ipr.2016.0972 [5] HE K M, SUN J, TANG X O. Single image haze removal using dark channel prior[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(12): 2341-2353. doi: 10.1109/TPAMI.2010.168 [6] ZHU Q S, MAI J M, SHAO L. A fast single image haze removal algorithm using color attenuation prior[J]. IEEE Transactions on Image Processing: A Publication of the IEEE Signal Processing Society, 2015, 24(11): 3522-3533. doi: 10.1109/TIP.2015.2446191 [7] YANG Y, WANG Z W. Haze removal: Push DCP at the edge[J]. IEEE Signal Processing Letters, 2020, 27: 1405-1409. doi: 10.1109/LSP.2020.3013741 [8] JU M Y, DING C, GUO Y J, et al. IDGCP: Image dehazing based on gamma correction prior[J]. IEEE Transactions on Image Processing, 2020, 29: 3104-3118. doi: 10.1109/TIP.2019.2957852 [9] LI B Y, PENG X L, WANG Z Y, et al. AOD-net: All-in-one dehazing network[C]//Proceedings of the IEEE International Conference on Computer Vision. Piscataway: IEEE Press, 2017: 4780-4788. [10] REN W Q, PAN J S, ZHANG H, et al. Single image dehazing via multi-scale convolutional neural networks with holistic edges[J]. International Journal of Computer Vision, 2020, 128(1): 240-259. doi: 10.1007/s11263-019-01235-8 [11] ZHENG Z R, REN W Q, CAO X C, et al. Ultra-high-definition image dehazing via multi-guided bilateral learning[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE Press, 2021: 16180-16189. [12] CHEN D D, HE M M, FAN Q N, et al. Gated context aggregation network for image dehazing and deraining[C]//Proceedings of the IEEE Winter Conference on Applications of Computer Vision. Piscataway: IEEE Press, 2019: 1375-1383. [13] BAI H R, PAN J S, XIANG X G, et al. Self-guided image dehazing using progressive feature fusion[J]. IEEE Transactions on Image Processing, 2022, 31: 1217-1229. doi: 10.1109/TIP.2022.3140609 [14] 杨燕, 梁小珍. 基于光幕约束和分段调整的去雾算法[J]. 光子学报, 2021, 50(5): 0510001.YANG Y, LIANG X Z. Dehazing algorithm based on atmospheric veil constraint and piecewise adjustment[J]. Acta Photonica Sinica, 2021, 50(5): 0510001 (in Chinese). [15] NAYAR S K, NARASIMHAN S G. Vision in bad weather[C]// Proceedings of the Seventh IEEE International Conference on Computer Vision. Piscataway: IEEE Press, 1999: 820-827. [16] 杨勇, 邱根莹, 黄淑英, 等. 基于改进大气散射模型的单幅图像去雾方法[J]. 北京航空航天大学学报, 2022, 48(8): 1364-1375.YANG Y, QIU G Y, HUANG S Y, et al. Single image dehazing method based on improved atmospheric scattering model[J]. Journal of Beijing University of Aeronautics and Astronautics, 2022, 48(8): 1364-1375 (in Chinese). [17] 金仙力, 张威, 刘林峰. 基于引导滤波和自适应容差的图像去雾算法[J]. 通信学报, 2020, 41(5): 27-36. doi: 10.11959/j.issn.1000-436x.2020078JIN X L, ZHANG W, LIU L F. Image defogging algorithm based on guided filtering and adaptive tolerance[J]. Journal on Communications, 2020, 41(5): 27-36 (in Chinese). doi: 10.11959/j.issn.1000-436x.2020078 [18] XIAO C B, ZHAO H Y, YU J, et al. Traffic image defogging method based on WLS[J]. Infrared and Laser Engineering, 2015, 44(3): 1080-1084. [19] MIN X K, ZHAI G T, GU K, et al. Quality evaluation of image dehazing methods using synthetic hazy images[J]. IEEE Transactions on Multimedia, 2019, 21(9): 2319-2333. doi: 10.1109/TMM.2019.2902097 [20] SUN W, WANG H, SUN C H, et al. Fast single image haze removal via local atmospheric light veil estimation[J]. Computers & Electrical Engineering: An International Journal, 2015, 46: 371-383. [21] 韩昊男, 钱锋, 吕建威, 等. 图像去雾方法质量评价[J]. 光学 精密工程, 2022, 30(6): 721-733. doi: 10.37188/OPE.20223006.0721HAN H N, QIAN F, LYU J W, et al. Image dehazing method quality assessment[J]. Optics and Precision Engineering, 2022, 30(6): 721-733 (in Chinese). doi: 10.37188/OPE.20223006.0721 -
下载:
点击查看大图
计量
- 文章访问数: 203
- HTML全文浏览量: 74
- PDF下载量: 1
- 被引次数: 0
