HALO-FREE DESIGN FOR RETINEX BASED REAL-TIME VIDEO ENHANCEMENT SYSTEM

Retinex becomes a prevailing theory in the field of image enhancement. In the literature, efforts on developing better illumination estimator have been made to alleviate the traditional halo artifacts existing on abrupt intensity changes. FPGA or DSP based real-time realization of Retinex algorithm is also studied. However, it is found that those real-time systems often suffer from a different kind of halo artifacts caused by inter-frame motion. That is because the algorithm is often realized with pipeline structure. In this paper, a complete halo-free design for Retinex based realtime video enhancement system is proposed. Separable bilateral filter, which is further optimized with a three-scale combined spatial Gaussian filter, is adopted to alleviate traditional halo artifacts. And an improved pipeline structure which adopts a motion estimation technique is proposed to restrain the motion caused halo artifacts. The overall design is analyzed and simulated. Experiments verify that the proposed design is able to achieve the goal of halo-free.

[1]  Michael Elad,et al.  Retinex by Two Bilateral Filters , 2005, Scale-Space.

[2]  Zia-ur Rahman,et al.  A multiscale retinex for bridging the gap between color images and the human observation of scenes , 1997, IEEE Trans. Image Process..

[3]  Michael Elad,et al.  A Variational Framework for Retinex , 2002, IS&T/SPIE Electronic Imaging.

[4]  Wen-Liang Hwang,et al.  Color image enhancement using retinex with robust envelope , 2009, 2009 16th IEEE International Conference on Image Processing (ICIP).

[5]  Sung-Jea Ko,et al.  Fast digital image stabilizer based on Gray-coded bit-plane matching , 1999, 1999 Digest of Technical Papers. International Conference on Consumer Electronics (Cat. No.99CH36277).

[6]  Lucas J. van Vliet,et al.  Separable bilateral filtering for fast video preprocessing , 2005, 2005 IEEE International Conference on Multimedia and Expo.

[7]  Glenn D. Hines,et al.  DSP Implementation of the Multiscale Retinex Image Enhancement Algorithm , 2004 .

[8]  A Hurlbert,et al.  Formal connections between lightness algorithms. , 1986, Journal of the Optical Society of America. A, Optics and image science.

[9]  Takao Onoye,et al.  An fpga implementation of real-time retinex video image enhancement , 2010, 2010 World Automation Congress.

[10]  Sung-Jea Ko,et al.  Fast digital image stabilizer based on Gray-coded bit-plane matching , 1999, IEEE Trans. Consumer Electron..

[11]  Roberto Manduchi,et al.  Bilateral filtering for gray and color images , 1998, Sixth International Conference on Computer Vision (IEEE Cat. No.98CH36271).

[12]  E. Land,et al.  Lightness and retinex theory. , 1971, Journal of the Optical Society of America.

[13]  Duan Hui-chuan,et al.  Image stabilization by combining gray-scale projection and block matching algorithm , 2009, 2009 IEEE International Symposium on IT in Medicine & Education.