ABSTRACT
The image registration by digital still cameras and video cameras requires color filters to be posed onto the photosensitive sensors (CCD or CMOS). The filters are arranged in patterns across the face of the image sensing array. The conventional color filter array (CFA) capture only one color component at each image pixel. The missing colors in the raw sensor data are interpolated by a process called CFA interpolation or demosaicing. Quality of the full-color reconstruction process is mostly relied on demosaicing method applied. Most of the current demosaicing methods are computationally expensive and often too slow for real-time scenarios. Many industrial applications require real-time and high quality demosaicing solutions, and quite often slow image reconstruction process is a real bottleneck. The purpose of this research is to present a comparative performance study of demosaicing algorithms on general-purpose GPUs. The experimental results of CUDA-based implementations of two state-of-the-art and widely applied in practice CFA algorithms are presented. The performance efficiency is assessed and analyzed by experimental studies on a set of real photographic test images on two general-purpose graphic cards. The obtained results demonstrated the benefit of exploiting the contemporary GPUs in speeding up the demosaicing process, especially for practical applications that need to meet real-time and high-speed video processing requirements combined with high quality of the full-color image reconstruction.
- P. Banerjee, & A. Dave, (2013). GPGPU Based Parallelized Client-Server Framework for Providing High Performance Computation Support. arXiv preprint arXiv:1505.05655.Google Scholar
- B. Bayer (1976), Color Imaging Array, Eastman Kodak Company Patent No.: 3.971.065.Google Scholar
- C., Kwan, B. Chou & Bell III, J. F. (2019). Comparison of Deep Learning and Conventional Demosaicing Algorithms for Mastcam Images. Electronics, 8(3), 308.Google ScholarCross Ref
- K. H. Chung, & Y. H. Chan, (2006). Color Demosaicing Using Variance of Color Differences. IEEE transactions on image processing, 15(10), 2944--2955. Google ScholarDigital Library
- Fastvideo (2019), Fastvideo SDK modules for CUDA image processing (ver. SDK.0.14.0.2.x64), Fastvideo.Google Scholar
- M. I. Faruqi, F. Ino, & K. Hagihara, (2012, July). Acceleration of variance of color differences-based demosaicing using CUDA. In 2012 International Conference on High Performance Computing & Simulation (HPCS) (pp. 503--510). IEEE.Google ScholarCross Ref
- P. Goorts, S. Rogmans, & P. Bekaert, (2012). Raw Camera Image Demosaicing using Finite Impulse Response Filtering on Commodity GPU Hardware using CUDA. INSTICC.Google Scholar
- Hasselblad©, Hasselblad sample images gallery 2019, {Online}. Available at https://www.hasselblad.com/learn/sample-images/. {Accessed 2019}.Google Scholar
- R. Langseth, V. R. Gaddam, H. K. Stensland, C. Griwodz, & P. Halvorsen, (2014, December). An Evaluation of Debayering Algorithms on GPU for Real-Time Panoramic Video Recording. In 2014 IEEE International Symposium on Multimedia (pp. 110--115). IEEE. Google ScholarDigital Library
- H. S. Malvar, L. W. He, & R. Cutler, (2004, May). High-quality linear interpolation for demosaicing of Bayer-patterned color images. In 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing (Vol. 3, pp. iii-485). IEEE.Google ScholarCross Ref
- D. Menon, S. Andriani, & G. Calvagno, (2007). Demosaicing with directional filtering and a posteriori decision. IEEE Transactions on Image Processing, 16(1), 132--141. Google ScholarDigital Library
- M. McGuire, (2008). Efficient, High-Quality Bayer Demosaic Filtering on GPUs. Journal of Graphics Tools, 13(4), 1--16.Google ScholarCross Ref
- N. G. Peter, (2009). NVIDIA's Fermi: The First Complete GPU Computing Architecture. A White Paper of NVIDIA.Google Scholar
- T. Wang, W. Guo, & J. Wei, (2018, August). An Optimization Scheme for Demosaicing Algorithm on GPU Using OpenCL. In CCF National Conference on Computer Engineering and Technology (pp. 142--152). Springer, Singapore.Google Scholar
- G. Zapryanov (2007) Interpolation Algorithms for Bayer Color Filter Array, Electrotechnique and Electronics (E+E) vol. 1, no. 2, pp. 68--73.Google Scholar
- G. Zapryanov, & I. Nikolova, (2008). Comparative Study of Demosaicing Algorithms for Bayer and Pseudo-Random Bayer Color Filter Array. In International Scientific Conference Computer Science.Google Scholar
- N. Zhang, J. C. Creput, H. Wang, C. Meurie, & Y. Ruichek, (2013, November). Partial demosaicing for stereo matching of CFA images on GPU and CPU. In 3rd International Conference on Advanced Communications and Computation, INFOCOMP (pp. 33--38).Google Scholar
Index Terms
- An Experimental Comparative Performance Study of Demosaicing Algorithms on General-purpose GPUs
Recommendations
Multi-directional Demosaicing for Digital Still Cameras
CSIE '09: Proceedings of the 2009 WRI World Congress on Computer Science and Information Engineering - Volume 07To obtain the full color image, at least three channels are required at each color location, however the image sensor obtains only the intensity of the brightness and cannot separate color information, three image sensors are required per pixel to ...
Demosaicing algorithms for area- and line-scan cameras in print inspection
Most color image sensors use color filter arrays (CFA). With this sensor design the captured information at each sensor pixel position is restricted to a specific spectral portion (typically red, green and blue bands). To obtain the missing color ...
Image demosaicing using content and colour-correlation analysis
Most digital cameras of today use a single CCD image sensor with alternating arrangement of red, blue and green colour filters in what is known as a Bayer pattern. To extract a full colour image, a demosaicing strategy has to be applied. In this paper ...
Comments