A Parametric Texture Model Based on Joint Statistics of Complex Wavelet Coefficients
International Journal of Computer Vision - Special issue on statistical and computational theories of vision: modeling, learning, sampling and computing, Part I
Self-similarity driven color demosaicking
IEEE Transactions on Image Processing
Color image demosaicking: An overview
Image Communication
Color plane interpolation using alternating projections
IEEE Transactions on Image Processing
Color demosaicking via directional linear minimum mean square-error estimation
IEEE Transactions on Image Processing
Adaptive Filtering for Color Filter Array Demosaicking
IEEE Transactions on Image Processing
| Hi-index | 0.08 |
We present an efficient approach for reconstructing full-color images from imagery data acquired using a color filter array (CFA). On the basis of our understanding of early visual processing in humans, we utilize a correlation among the multi-scale directional-anisotropy statistics observed in natural images in order to estimate the missing high-spatial-frequency chromatic signals. The directional anisotropy in the high-spatial-frequency component is efficiently and robustly estimated according to a mixture of anisotropy statistics derived from lower-frequency components. We show that, in spite of its simple implementation, our proposed method gives an excellent numerical performance as well as a perceptually natural output. The present approach is expected to provide a promising methodology for the on-line processing of recently rising high-resolution image data acquired using a single-sensor CFA.