초록 ▼

For processing image data generated for a color filter array, classifiers are generated from the image data. A predetermined color component is determined by directional-interpolating the image data depending on the classifiers. Chrominance components are generated from linear-interpolated image dat

For processing image data generated for a color filter array, classifiers are generated from the image data. A predetermined color component is determined by directional-interpolating the image data depending on the classifiers. Chrominance components are generated from linear-interpolated image data and the directional-interpolated predetermined color component. The chrominance components are directional-interpolated, and color components are extracted from the directional-interpolated chrominance components.

대표청구항 ▼

What is claimed is: 1. A method of processing image data generated according to a color filter array, comprising: generating horizontal and vertical classifiers from the image data; directional-interpolating a predetermined color component from the image data depending on the classifiers to generat

What is claimed is: 1. A method of processing image data generated according to a color filter array, comprising: generating horizontal and vertical classifiers from the image data; directional-interpolating a predetermined color component from the image data depending on the classifiers to generate a directional-interpolated color component, wherein the predetermined color component is determined from a selected one of a row of the image data or a column of the image data depending on which of the horizontal classifier and the vertical classifier is greater; and generating chrominance components from the directional-interpolated predetermined color component and image data for at least one other color component aside from the predetermined color component, wherein the directional-interpolated color component has a value of a horizontal convolution current if the vertical classifier is greater than the horizontal classifier, a value of a vertical convolution current if the vertical classifier is less than the horizontal classifier, and an average of the values of the horizontal and vertical convolution currents if the vertical classifier is equal to the horizontal classifier. 2. The method of claim 1, wherein the color filter array has a Bayer color pattern. 3. The method of claim 2, wherein the predetermined color component is for a green color. 4. The method of claim 3, wherein the image data is linear-interpolated data for data generated from a color filter array and the at least one other color component includes linear-interpolated image data of red and blue colors. 5. The method of claim 1, wherein the step of generating the classifiers includes: calculating the horizontal classifier for a current pixel using the image data for the current pixel and additional pixels to a left and right of the current pixel; and calculating the vertical classifier for the current pixel using the image data for the current pixel and additional pixels to a top and bottom of the current pixel. 6. The method of claim 5, further comprising: adjusting the horizontal classifier by averaging horizontal classifiers of the current pixel and adjoining pixels, and adjusting the vertical classifier by averaging vertical classifiers of the current pixel and adjoining pixels, when a difference between the horizontal and vertical classifiers for the current pixel is less than a threshold. 7. The method of claim 1, further comprising: calculating a horizontal convolution value from the row of the image data that is selected for the directional-interpolated predetermined color component if the vertical classifier is greater than the horizontal classifier; calculating a vertical convolution value from the column of the image data that is selected for the directional-interpolated predetermined color component if the vertical classifier is less than the horizontal classifier; and averaging the horizontal and vertical convolution values for the directional-interpolated predetermined color component if the vertical classifier is substantially equal to the horizontal classifier. 8. The method of claim 1, further comprising: recalculating the classifiers using the directional-interpolated predetermined color component. 9. The method of claim 1, further comprising: directional-interpolating the chrominance components depending on the classifiers; and extracting at least one color component of a pixel using the directional-interpolated chrominance components of the pixel. 10. An apparatus for processing image data generated according to a color filter array, comprising: a classifier calculator for generating horizontal and vertical classifiers from the image data; a first directional interpolator for directional-interpolating a predetermined color component from the image data depending on the classifiers to generate a directional-interpolated color component, wherein the predetermined color component is determined from a selected one of a row of the image data or a column of the image data depending on which of the horizontal classifier and the vertical classifier is greater; and a color difference calculator for generating chrominance components from the directional-interpolated predetermined color component and image data for at least one other color component aside from the predetermined color component, wherein the directional-interpolated color component has a value of a horizontal convolution current if the vertical classifier is greater than the horizontal classifier, a value of a vertical convolution current if the vertical classifier is less than the horizontal classifier, and an average of the values of the horizontal and vertical convolution currents if the vertical classifier is equal to the horizontal classifier. 11. The apparatus of claim 10, wherein the color filter array has a Bayer color pattern. 12. The apparatus of claim 11, wherein the predetermined color component is for a green color. 13. The apparatus of claim 12, wherein the image data is linear-interpolated data for data generated from a color filter array and the at least one other color component includes linear-interpolated image data of red and blue colors. 14. The apparatus of claim 10, wherein the classifier calculator calculates the horizontal classifier for a current pixel using the image data for the current pixel and additional pixels to a left and right of the current pixel, and calculates the vertical classifier for the current pixel using the image data for the current pixel and additional pixels to a top and bottom of the current pixel. 15. The apparatus of claim 14, wherein the classifier calculator adjusts the horizontal classifier by averaging horizontal classifiers of the current pixel and adjoining pixels, and adjusts the vertical classifier by averaging vertical classifiers of the current pixel and adjoining pixels, when a difference between the horizontal and vertical classifiers for the current pixel is less than a threshold. 16. The apparatus of claim 10, wherein the first directional interpolator calculates a horizontal convolution value from the row of the image data that is selected for the directional-interpolated predetermined color component if the vertical classifier is greater than the horizontal classifier, calculates a vertical convolution value from the column of the image data that is selected for the directional-interpolated predetermined color component if the vertical classifier is less than the horizontal classifier, and averages the horizontal and vertical convolution values for the directional-interpolated predetermined color component if the vertical classifier is substantially equal to the horizontal classifier. 17. The apparatus of claim 10, wherein the classifier calculator recalculates the classifiers using the directional-interpolated predetermined color component. 18. The apparatus of claim 10, further comprising: a second directional interpolator for directional-interpolating the chrominance components depending on the classifiers; and a color component generator for extracting at least one color component of a pixel using the directional-interpolated chrominance components of the pixel.