摘要
为了提高图像去雾效果,提出一种改进暗通道(IDCP)算法,通过缩小图像结合3个暗通道线性拟合方法对透射率进行精确计算,在大气值一定区间内利用四叉树细分算法求取图像的大气光值最终估计值,为避免图像饱和度偏低进行了颜色补偿。实验仿真显示该算法去雾后的图像在视觉上更加清新自然,对不同的图像的客观评价值优于其他算法。
In order to improve the effect of image dehazing,an improved dark channel prior( IDCP) algorithm was proposed. Transmittance was calculated accurately with narrow image and three dark channel prior linear fitting method,atmospheric light was estimated by four fork tree algorithm within a certain interval,color was compensated in order to avoid the image lower saturation. Experimental results show that IDCP is more refreshing and natural in vision,and the comprehensive evaluation value of different images is better than other algorithms.
引文
[1]赵政达.关于暗通道先验图像去雾算法的改进[D].长春:吉林大学,2014.Zhao Z D. Image dehazing based on improvement dark channel prior algorithm[D]. Changchun:Jilin University,2014.
[2] He K M,Sun J,Tang X O. Single image haze removal using dark channel prior[J]. Pattern Analysis and Machine Intelligence,2011,33(12):2341-2353.
[3] Gibson K B,Vo D T, Nguyen T Q, et al. An investigation of dehazing effects on image and video coding[J]. IEEE Trans on Image Process,2012,21(2):662–673.
[4] Wang J B,He N,Zhang L L. Single image dehazing with a physical model and dark channel prior[J].Neurocomputing,2015,149:718-728.
[5] Tan R T. Visibility in bad weather from a single image[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Anchoragc,AK,USA,2008:1-8.
[6] Kratz L,Nishino K. Factorizing Scene Albedo and Depth from a Single Foggy Image[J]. IEEE International Conference on Computer Vision,2010,30(2):1701-1708.
[7]陶海威,王毅,沈燕飞,等.基于暗通道先验的图像去雾算法改进研究[J].软件导刊,2016,15(5):30-34.Tao H W,Wang Y,Shen Y F,et al. Improved algorithm of image Dehazing based on dark channel priors[J].Software Guide,2016,15(5):30-34.
[8]陈勇,郝裕斌,张开碧.结合K均值聚类的暗通道去雾改进算法[J].半导体光电,2016,37(6):876-881.Chen Y,Hao Y B,Zhang K B. Improvement of Haze Removal Method Combining Dark Channel Priority with K-means Clustering Algorithm[J]. Semiconductor Optoelectronics,2016,37(6):876-881.
[9]李校林,王屈桥,燕历科.基于改进暗通道先验模型的遥感图像去雾算法[J].电视技术,2017,41(1):14-17.Li X L,Wang Q Q,Yan L K. UAV remote sensing image de-hazing algorithm based on improved dark channel priormodel[J]. Video engineering, 2017,41(1):14-17.
[10]刘哲,黄文准,黄世奇.一种快速单幅图像去雾新方法[J].计算机系统应用,2017,26(1):16-23.Liu Z,Huang W Z,Huang S Q. New Fast Single Image Dehazeing Method[J]. Computer Systems&Applications,2017,26(1):16-23.
[11]刘万军,赵庆国.基于颜色饱和度的快速图像去雾研究[J].计算机工程与应用,2017,53(8):209-215.Liu W J,Zhao Q G. Fast image dehaze study based on color saturation[J]. Computer Engineering and Applications,2017,53(8):209-215.
[12]朱奇光,张兴家,陈卫东,等.基于颜色矩的改进尺度不变特征变换的移动机器人定位算法[J].计量学报,2016,37(2):118-122.Zhu Q G,Zhang X J,Chen W D,et al. An Algorithm Research for Mobile Robot Localization Based on the Improved Scale Invariant Feature Transform of Color Moment[J]. Acta Metrologica Sinica,2016,37(2):118-122.
[13]黄素霞.单幅图像中值暗通道先验去雾的改进算法[J].光电子·激光,2015,26(8):1611-1617.Huang S X. Improvement algorithm for single image haze removing by median dark channel prio[J].Journal of Optoelectronics and lasers,2015(8):1611-1617.
[14]张航,潘浩,徐军,等.基于色调单维度的相关色温计算[J].计量学报,2017,38(1):56-59.Zhang H,Pan H,Xu J,et al. Calculation of Correlated Color Temperature by Single-dimension of Hue[J]. Acta Metrologica Sinica,2017,38(1):56-59.