True Color Image Enhanced by Wavelet Illumination-Reflection Model

Abstract:

When a true color image is enhanced, the same hue should be kept and the useful details should be strengthened in the image enhanced. According to the imaging principle, wavelet illumination-reflection method is proposed. Firstly, true color image is transformed from RGB space to HSV space. Secondly, the saturation is attenuated by CLAHE. Thirdly, the value is decomposed into illumination and reflection by wavelet illumination-reflection model. The details of the reflection are strengthened. The dynamic range of the illumination is reduced in order to enhance the image. Experiments and analysis show that the enhancement algorithm based on wavelet illumination-reflection model is obviously better than multi-scales Retinexs with color restoration (MSRCR).

• Periodical: Applied Mechanics and Materials (Volumes 397-400)
• Pages: 1500-1505
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.397-400.1500
• Online since: September 2013
• Authors: Jie Xiong, Lei Lu
• Keywords: HSV Space, MSRCR, RGB Space, Saturation, Stationary Wavelet Transform, True Color Image, Value, Wavelet Illumination-Reflection Model

References

[1] Xiong. Jie, Han. Li-na, Geng. guo-hua and Zhou. Ming-quan : Digital medical image enhanced by similar method of Retinex [J] Computer Engineering and Applications. Vol 45-24. (2009), pp.14-16 (in Chinese).

[2] Xiong. Jie, Han. Li-na, Geng. guo-hua and Zhou. Ming-quan Using HSV space real-color image enhanced by homomorphic filter in two channel [J] Computer Engineering and Applications. Vol 45-27. (2009), pp.18-20 (in Chinese).

DOI: 10.1109/iccsit.2009.5234693

[3] Jobson.D. J, Rahmam. Z, Woodell.G. A Diffusion Properties and performance of a center/surround retinex, [J] IEEE Trans on Image Processing: Special Issue on Color Processing Vol 6-7. (1997), pp.451-462.

DOI: 10.1109/83.557356

[4] Wang. Yan-chen, LI. Shu-jie, Huang. Lian-qing Enhancement of radiography based multiscale Retinex [J]. Optics and Precision Engineering Vol 14-1. (2006), pp.70-77 (in Chinese).

[5] Xiong. Jie, Real color image enhancement based on frequency domain, wavelet transform and neural network [D] Northwest University doctoral thesis (2010) (in Chinese).

[6] Jobson.D. J, Rahmam. Z, Woodell.G. A The statistics of visual representation [C] Proceedings of SPIE Visual Information Processing. Washington: SPIE Press, 2002, pp.25-35.

[7] Fu. Zu-yun, Information Theory [M] Beijing: Electronics Industry Press, 2001 (in Chinese).

[8] Yan. Jin-wen, Digital Image Processing [M] Beijing: National Defense Industry Press , 2007. p.95 (in Chinese).

[9] Zuiderveld, Karel . Contrast Limited Adaptive Histograph Equalization [J] Graphic Gems IV. San Diego: Academic Press Professional, 1994. 474–485.

DOI: 10.1016/b978-0-12-336156-1.50061-6

[10] Xiong. Jie, Feng. Xiao-qiang, Geng. guo-hua and Zhou. Ming-quan Real Color Image Enhanced and Denoised by Stationary Wavelet Transform [J] Computer Science. Vol . 36 No. 11. (2009), pp.254-257(in Chinese).

DOI: 10.1109/iccsit.2009.5234693