A method of reducing color aliasing artifacts from a color digital image having color pixels including providing luminance and chrominance signals from the color digital image; downsampling the luminance and chrominance signals; using such downsampled luminance and chrominance signals to separate the image into textured and nontextured regions having boundaries; cleaning the downsampled chrominance signals in the textured regions in response to the boundaries of the textured region and the downsampled chrominance signals; cleaning the downsampled chrominance signals in the nontextured regions in response to the downsampled chrominance signals; upsampling the downsampled noise-cleaned chrominance signals; and using the luminance and upsampled noise-cleaned chrominance signals to provide a color digital image having reduced color aliasing artifacts.

Demosaicing of graphical content is provided. In an illustrative implementation a demosaicing engine executing one or more demosaicing algorithms is employed to operate on graphical content to provide better quality and higher resolution images. In operation, the demosaicing engine operates in two modes, a training/learning mode, and a run time mode. During training, training-images are analyzed to generate a codebook of mosaic filter table entries, such that each table entry has an associated list of similar training pixel blocks and their associated filters. During run time, a run-time image is broken into pixel blocks. Each pixel block is then compared with the entries of the codebook to find the closest match filter. The list associated with the entry is then processed using a least-squares algorithm to locate the optimal mosaic filter. As a result, higher resolution is achieved without requiring more pixels.

Demosaicing of graphical content is provided. In an illustrative implementation a demosaicing engine executing one or more demosaicing algorithms is employed to operate on graphical content to provide better quality and higher resolution images. In operation, the demosaicing engine operates in two modes, a training/learning mode, and a run time mode. During training, training-images are analyzed to generate a codebook of mosaic filter table entries, such that each table entry has an associated list of similar training pixel blocks and their associated filters. During run time, a run-time image is broken into pixel blocks. Each pixel block is then compared with the entries of the codebook to find the closest match filter. The list associated with the entry is then processed using a least-squares algorithm to locate the optimal mosaic filter. As a result, higher resolution is achieved without requiring more pixels.

A spatial transformation methodology provides a new image interpolation scheme, or analyzes an already existing one. Examples of spatial operations include but are not limited to, demosaicing, edge enhancement or sharpening, linear filtering, and non-linear filtering. A demosaicing operation is described herein, although the scheme is applied generally to spatial transformation operations. The spatial transformation methodology includes detailed expressions for the noise covariance after a spatial operation is performed for each of the three color channels, red, green, and blue. A color filter array is in the form of a Bayer pattern and demosaicing is performed using a 4-neighbor bilinear interpolation. Using lattice theory, the spatial transformation methodology predicts noise covariance after demosaicing in terms of the input noise covariance and an autocorrelation function of the image is determined for a given selectable number of shifts.

A method of processing a mosaic digital image includes simultaneously performing edge-sensitive denoising and color interpolation on a first color plane of the digital image. The interpolation fills in missing pixel information in the first color plane.