模糊域内基于Retinex的雾霾图像增强算法

计算机科学 ›› 2015, Vol. 42 ›› Issue (Z11): 183-188.

模糊域内基于Retinex的雾霾图像增强算法

贾伟,刘彦宾,刘炜,葛耿育,苏文丽,樊里略

遵义师范学院计算机与信息科学学院遵义563002,遵义师范学院计算机与信息科学学院遵义563002,遵义师范学院计算机与信息科学学院遵义563002;贵州大学计算机科学与技术学院贵阳563003,遵义师范学院计算机与信息科学学院遵义563002,遵义师范学院计算机与信息科学学院遵义563002,遵义师范学院计算机与信息科学学院遵义563002

出版日期:2018-11-14 发布日期:2018-11-14
基金资助:
本文受贵州省科技厅、遵义市科技局、遵义师范学院联合基金项目(黔科合J字LKZS[2014]23号,黔科合J字LKZS[2014]08号),遵义师范学院教学科研项目(13-08,13-06)资助

Foggy and Hazy Image Enhancement Algorithm Based on Retinex in Fuzzy Field

JIA Wei, LIU Yan-bin, LIU Wei, GE Geng-yu, SU Wen-li and FAN Li-lue

Online:2018-11-14 Published:2018-11-14

摘要/Abstract

摘要： 现有Retinex图像增强算法在使用过程中往往会产生轻微的光晕现象,在图像清晰度、细节、保真性、适用范围等方面存在诸多不足,在模糊域内,提出的基于Retinex的雾霾图像增强算法可有效克服此类不足。首先利用自适应多阈值算法对图像进行分块,并确定分块区域的最佳渡越点；然后采用线性隶属度函数将图像像素值变换为模糊域,通过渡越点计算提出的模糊双曲正切函数的关联参数,对图像的各个分块区域应用Retinex算法进行非线性图像增强,同时对增强结果进行模糊双曲正切调整；最后采用线性加权和线性逆变换的方法恢复原图像。通过对大量图像的对比实验发现:传统处理方法产生的光晕现象得到了抑制,图像清晰度、细节、保真性以及对比度等处理效果改善明显,算法适用范围更广。

关键词: 模糊域,雾霾图像增强,Retinex算法,模糊双曲正切

Abstract: The traditional Retinex algorithm always products “halo” effect in fog and haze image enhancement,and has several shortcomings,such as the poor image exposure,image sharpness,image details,image fidelity and so on.In order to overcome these shortcomings of the traditional algorithm,we proposed a foggy and hazy image enhancement algorithm based on Retinex in the fuzzy field.Firstly,the original image was classified into several blocks using the proposed adaptive multi-threshold algorithm,and then the optimal crossover points of the image blocking areas were computed.Secondly,we used a novel linear membership function to map the image pixel value into fuzzy domain,then computed the correlation parameter of the fuzzy hyperbolic tangent function by crossover points,used the Retinex algorithm to perform no-linear image enhancement,and used the fuzzy hyperbolic tangent functions to adjust the enhancement result.Finally,the method of superposition of the linear was used to map the enhancement result into original image domain.The experimental result shows that the “halo” effect is suppressed,the proposed method plays a better role in the image sharpness,image details and image fidelity,and it has more widely applicability.

Key words: Fuzzy field,Foggy and hazy image enhancement,Retinex algorithm,Fuzzy hyperbolic tangent function

贾伟,刘彦宾,刘炜,葛耿育,苏文丽,樊里略. 模糊域内基于Retinex的雾霾图像增强算法[J]. 计算机科学, 2015, 42(Z11): 183-188. https://doi.org/

JIA Wei, LIU Yan-bin, LIU Wei, GE Geng-yu, SU Wen-li and FAN Li-lue. Foggy and Hazy Image Enhancement Algorithm Based on Retinex in Fuzzy Field[J]. Computer Science, 2015, 42(Z11): 183-188. https://doi.org/

参考文献

[1] Koschmieder H.Theorie der horizontalen sichtweite[J].Beitr.Phys.Freien Atm.,1924,12:171-181
[2] Narasimhan S G,Nayar S K.Vision and the atmosphere[J].International Journal of Computer Vision,2002,48(3):233-254
[3] Schechner Y Y,Narasimhan S G,Nayar S K.Instant dehazing of images using polarization[C]∥Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition,2001(CVPR 2001).IEEE,2001
[4] He K,Sun J,Tang X.Single image haze removal using darkchannel prior[J].IEEE Transactionson Pattern Analysis and Machine Intelligence,2011,33(12):2341-2353
[5] Lan X,Zhang L,Shen H,et al.Single image haze removal considering sensor blur and noise[J].EURASIP Journal on Advances in Signal Processing,2013,2013(1):1-13
[6] Serikawa S,Lu H.Underwater image dehazingusing joint trila-teral filter[J].Computers & Electrical Engineering,2014,40(1):41-50
[7] Land E H,McCann J.Lightness and Retinex theory[J].JOSA,1971,61(1):1-11
[8] Jobson D J,Rahman Z U,Woodell G A.Properties andperformance of center/surround Retinex[J].IEEE Transactions on Image Processing,1997,6(3):451-462
[9] Rahman Z U,Jobson D J,Woodell G A.Multi-scale Retinex for color image enhancement[C]∥Proceedings.International Conference on Image Processing,1996.IEEE,1996,3:1003-1006
[10] Rahman Z,Jobson D J,Woodell G A,et al.Image enhancement,image quality,and noise[C]∥International Society for Optics and Photonics Optics & Photonics 2005.2005
[11] 张新龙,汪荣贵,张璇,等.雾天图像增强计算模型及算法[J].中国图象图形学报,2011,16(8):1359-1368
[12] 李菊霞,余雪丽.雾天条件下的多尺度 Retinex图像增强算法[J].计算机科学,2013,40(3):299-301
[13] Pal S K,King R A.Image enhancement using fuzzy set[J].Electronics letters,1980,16(10):376-378
[14] Pal S K,King R A.On edge detection of X-ray images using fuzzy sets[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1983(1):69-77
[15] Pal S K,King R A.Image enhancement usig smoothing with fuzzy sets[J].IEEE Transactions on Systems,Man and Cybernetics-Part A:Systems and Humans,1981,11(7):494-501
[16] Otsu N.A threshold selection method from gray levelhistogram[J].IEEE Transactions on System,Man and Cybernetics,1979,9(1):62-66
[17] 刘艳,赵英良.Otsu 多阈值快速求解算法[J].计算机应用,2011,1(12):3363-3365
[18] Cheng H D,Sun Y.A hierarchical approach to color image segmentation using homogeneity[J].IEEE Transactions on Image Processing,2000,9(12):2071-2082
[19] Zhang H,Wang Z,Liu D.Chaotifying fuzzy hyperbolic modelusing adaptive inverse optimal control approach[J].International Journal of Bifurcation and Chaos,2004,14(10):3505-3517
[20] Wang Y,Zhang H.Robust H_∞ Control of Stochastic Hyperbolic Tangent Model [J].Journal of Northeastern University(Natural Science),2008,29(1):2
[21] 门绍雄,谭冠荣.图象的双曲正切函数变换[J].仪器仪表学报,1984,3:104-108
[22] 薛向阳,罗航哉.一种新的颜色相似度定义及其计算方法[J].计算机学报,1999,22(9):918-922

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed