ALGORITHMS FOR IMAGE RECONSTRUCTION BASED ON BAYER MOSAIC
DOI:
https://doi.org/10.24867/29RB01TesanovicKeywords:
Bayer mosaic, guided image filtering, residual interpolation, CNNs, demosaickingAbstract
The paper discusses the process of color image generation using matrix image sensors with Bayer mosaic. The image processing begins with measuring the amount of light in the red, green and blue part of the spectrum. Instead of a light beam splitter that would project the image onto three different photosensitive elemnts, in the case of Bayer mosaic, a series of color filters with a specific spatial arrangement on the matrix sensor is used. The method of interpolation of missing spectral measurements (demosaicking) and the sequence of operations such as denoising and spatial resolution enhancement are not arbitrary and significantly affect the quality of the reconstructed image. Interpolation techniques based on Hamilton’s interpolation formulas and guided image filtering using residual interpolation are analyzed in the paper. Experimental comparison of reconstruction results with techniques based on convolutional neural networks (CNN) identified the most effective approaches image quality improvement and demosaicking.
References
[2] Qian, G., Wang, Y., Gu, J., Dong, C., Heidrich, W., Ghanem, B., Ren, J. S. (2023). Rethinking learning-based demosaicing, denoising, and super-resolution pipeline. IEEE International Conference on Computational Photography (ICCP), pp. 1-12
[3] Kiku, D., Monno, Y., Tanaka, M., Okutomi, M. (2013). Residual interpolation for color image demosaicking. IEEE Transactions on Image Processing, Vol. 22, No. 11, pp. 4270-4280.
[4] Kiku, D., Monno, Y., Tanaka, M., Okutomi, M. (2014). Minimized-Laplacian residual interpolation for color image demosaicking. IS&T/SPIE Electronic Imaging (EI), Digital Photography X, Vol. 9023, pp. 90230L-1-8.
[5] Monno, Y., Kiku, D., Tanaka, M., Okutomi, M. (2015). Adaptive residual interpolation for color image demosaicking. IEEE International Conference on Image Processing (ICIP), pp. 3861-3865
[6] Tan, R., Zhang, K., Zuo, W., Zhang, L. (2017). Color image demosaicking via deep residual learning. IEEE Int. Conf. Multimedia Expo (ICME), Vol. 2, No. 4, pp. 1-6.
[7] He, K., Sun, J., Tang, X. (2013). Guided image filtering. IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 35, No. 6, pp. 1397-1409.
[8] Pekkucuksen, I., Altunbasak, Y. (2010). Gradient based threshold free color filter array interpolation, IEEE International Conference on Image Processing (ICIP), Hong Kong, pp. 1377-1380.
[9] Hamilton, J.F., Adams J.E. (1997). Adaptive color plan interpolation in single sensor color electronic camera, U.S. Patent 5,629,734, 1997.
