A Fast Image Enhancement Algorithm Based on Fusion of Different Color Spaces
-
摘要: 针对现有Retinex算法中存在的色彩失真、噪声放大及光晕伪影现象等问题,本文提出了一种基于Retinex理论的改进算法. 该算法首先在HSV空间对亮度分量V通道进行增强处理,同时在拉伸得到的对数域反射分量至一定的动态范围时(本文是0~255),引入增强调整因子,调整不同亮度值的增强程度来避免噪声放大及色彩失真现象;然后在RGB空间,通过分析光晕产生的原因,提出一种改进的高斯滤波器来消除光晕现象,并在计算反射分量时,通过参数调整图像颜色的保真度. 最后,对上述两种不同颜色空间的处理结果进行加权平均作为算法的最终输出. 实验结果表明,针对不同光照条件下的图像,1)该算法可以明显地改善光晕伪影现象;2)无色彩失真、噪声放大等问题;3)效果和效率优于带色彩恢复的多尺度Retinex算法(Multi-scale retinex with color restoration,MSRCR)及其他对比算法.
-
关键词:
- Retinex理论 /
- 带色彩恢复的多尺度Retinex (MSRCR) /
- 光晕伪影 /
- 图像增强
Abstract: A fast image enhancement algorithm based on fusion of different color spaces is proposed to overcome the problems of color distortion, noise amplification, and halo artifacts. Firstly, a single scale retinex is used in channel V of the HSV color space. An enhancement adjustment factor is introduced, when we stretch the calculated reflection in the logarithmic domain into a dynamic range, i.e., 0~255. It is used to adjust the enhancement for different pixels. Thus noise amplification and color distortion can be effectively avoided. Then, the improved Gaussian filter is given in the RGB color space by analyzing the cause of the halo effect. This step can eliminate the halo artifact. And a parameter is adopted to keep the color natural of the image when the reflection is calculated. Finally, the weighted average of the outputs of the above two color spaces is taken as the final output of our algorithm. The experiment results show that for images with different lighting conditions, 1) the outputs of our algorithm is free from the halo artifacts; 2) there are no color distortion and noise amplification problems; 3) the quality and the efficiency of the algorithm are superior to the multi-scale retinex algorithm (MSRCR) and other comparison algorithms. -
[1] Jobson D J, Rahman Z, Woodell G A. Properties and performance of a center/surround retinex. IEEE Transactions on Image Processing, 1997, 6(3): 451-462 [2] Jobson D J, Rahman Z, Woodell G A. A multiscale retinex for bridging the gap between color images and the human observation of scenes. IEEE Transactions on Image Processing, 1997, 6(7): 965-976 [3] Meylan L, Susstrunk S. High dynamic range image rendering with a retinex-based adaptive filter. IEEE Transactions on Image Processing, 2006, 15(9): 2820-2830 [4] Chi Jian-Nan, Zhang Chuang, Zhang Zhao-Hui, Wang Zhi-Liang. Image enhancement based on anti-symmetrical biorthogonal wavelet recontruction. Acta Automatica Sinica, 2010, 36(4): 475-487 (迟健男, 张闯, 张朝晖, 王志良. 基于反对称双正交小波重构的图像增强方法. 自动化学报, 2010, 36(4): 475-487) [5] Choi D H, Jang I H, Kim M H, Kim N C. Color image enhancement based on single-scale retinex with a JND-based nonlinear filter. In: Proceedings of IEEE International Symposium on Circuits and Systems. New Orleans, USA: IEEE, 2007. 3948-3951 [6] Zhang Y Q, Ding Y, Xiao J S, Liu J Y, Guo Z M. Visibility enhancement using an image filtering approach. EURASIP Journal on Advances in Signal Processing, 2012, 2012(1): 1-6 [7] Li B, Wang S H, Geng Y B. Image enhancement based on Retinex and lightness decomposition. In: Proceedings of the 18the IEEE International Conference on Image Processing. Brussels: IEEE, 2011. 3417-3420 [8] Wang Rong-Gui, Zhu Jing, Yang Wan-Ting, Fang Shuai, Zhang Xin-Tong. An improved local multi-scale Retinex algorithm based on illuminance image segmentation. Acta Electronica Sinica, 2010, 38(5): 1181-1186(汪荣贵, 朱静, 杨万挺, 方帅, 张新彤. 基于照度分割的局部多尺度Retinex算法. 电子学报, 2010, 38(5): 1181-1186) [9] Liu Wei-Hua, Sui Qing-Mei. A hue preserving MSR algorithm of image enhancement. Acta Photonica Sinica, 2011, 40(4): 642-646(刘伟华, 隋青美. 基于色调恒定的MSR图像增强算法. 光子学报, 2011, 40(4): 642-646) [10] Yu Jing, Li Da-Peng, Liao Qing-Min. Color constancy-based visibility enhancement of color images in low-light conditions. Acta Automatica Sinica, 2011, 37(8): 923-931(禹晶, 李大鹏, 廖庆敏. 基于颜色恒常性的低照度图像视见度增强. 自动化学报, 2011, 37(8): 923-931) [11] Hanumantharaju M C, Ravishankar M, Rameshbabu D R, Ramachandran S. Color image enhancement using multiscale Retinex with modified color restoration technique. In: Proceedings of the 2011 Second International Conference on Emerging Applications of Information Technology. Kolkata: IEEE, 2011. 93-97 [12] Mukherjee J, Mitra S K. Enhancement of color images by scaling the DCT coefficients. IEEE Transactions on Image Processing, 2008, 17(10): 1783-1794 [13] Yu J, Kim Y, Lee N, Tchoi Y S, Hahn H S. Effective color correction method employing HSV color model. Journal of Measurement Science and Instrumentation, 2012, 3(1): 39-45 [14] Wang W, Li B, Zheng J, Xian S. A fast multi-scale retinex algorithm for color image enhancement. In: Proceedings of International Conference on Wavelet Analysis and Pattern Recognition. Hong Kong, China: IEEE, 2008. 80-85 [15] Wang Ke-Jun, Xiong Xin-Yan, Ren Zhen, Fu Bin. MSR image enhancement based on window irrelevant mean filter. Journal of Harbin Institute of Technology, 2010, 42(9): 1467-1471(王科俊, 熊新炎, 任桢, 付斌. 基于窗口无关均值滤波的MSR图像增强. 哈尔滨工业大学学报, 2010, 42(9): 1467-1471) [16] Rahman Z, Jobson D J, Woodell G A. Retinex processing for automatic image enhancement. Journal of Electronic Imaging, 2004, 13(1): 100-110
点击查看大图
计量
- 文章访问数: 2070
- HTML全文浏览量: 174
- PDF下载量: 1226
- 被引次数: 0