Apparatus and method for image edge-preserving smoothing
Provided is an image processing apparatus including: a frequency value calculation section that allocates each pixel of an input image to any of respective partial regions obtained by dividing the entirety of a possible range of a luminance value into units in a luminance direction on the basis of the luminance value thereof, and allocates one pixel of the input image to partial regions when calculating frequency values representing the number of pixels allocated to the partial regions with respect to the respective partial regions, to update the frequency values of the partial regions; a characteristic value calculation section that calculates a characteristic value representing a characteristic of the partial region; and a weighted product-sum section that performs edge-preserving smoothing on the input image by weighting and averaging the characteristic values in accordance with a distance in the luminance direction, using the calculated frequency value and the calculated characteristic value.
1. An image processing apparatus comprising:
one or more processors operable to:
allocate each pixel of an input image to a plurality of partial regions, wherein each of the plurality of partial regions corresponds to a predetermined range of luminance values;
determine a number of pixels of the input image, allocated to each of the plurality of partial regions, up to a decimal precision of a predetermined number of bits;
determine a characteristic value corresponding to each of the plurality of partial regions; and
perform edge-preserving smoothing on the input image by weighting and averaging characteristic values in accordance with a distance in luminance direction, using the number of pixels of the input image allocated to each of the plurality of partial regions and the determined characteristic value.
2. The image processing apparatus according to claim 1 , wherein the one or more processors are operable to:
determine a product of a luminance value of each pixel of the input image by corresponding weights of each the plurality of partial regions; and
update the characteristic value with the product of the luminance value and the weight corresponding to each pixel of each of the plurality of partial regions.
3. The image processing apparatus according to claim 1 , wherein the one or more processors are operable to calculate weights corresponding to each of the plurality of partial regions in accordance with a predetermined weighting function based on the number of pixels of the input image allocated to each of the plurality of partial regions, up to the decimal precision of the predetermined number of bits.
4. The image processing apparatus according to claim 1 , wherein the one or more processors are operable to:
decimate pixels of the input image; and
determine weights corresponding to each of the plurality of partial regions with respect to the decimated pixels of the input image.
5. The image processing apparatus according to claim 1 , wherein the one or more processors are operable to decimate the pixels of the input image at a decimation rate corresponding to the decimal precision of the number of pixels of the input image allocated to each of the plurality of partial regions.
6. The image processing apparatus according to claim 1 , wherein the one or more processors are operable to:
determine a spatial pattern representing contents of the input image,
decimate the pixels of the input image at a decimation rate in accordance with an analysis result of the determined spatial pattern.
7. The image processing apparatus according to claim 1 , wherein the one or more processors are operable to:
determine spatial patterns for each of the plurality of partial regions of the input image; and
decimate the pixels of the input image, for each partial region, at a decimation rate in accordance with the analysis result of a corresponding spatial pattern.
8. An image processing method in an image processing apparatus, the image processing method comprising:
allocating each pixel of an input image to a plurality of partial regions, wherein each of the plurality of partial regions corresponds to a predetermined range of luminance value;
determining a number of pixels of the input image, allocated to each of the plurality of partial regions, up to a decimal precision of a predetermined number of bits;
determining a characteristic value corresponding to each of the plurality of partial regions; and
performing edge-preserving smoothing on the input image by weighting and averaging characteristic values in accordance with a distance in luminance direction, using the number of pixels of the input image allocated to each of the plurality of partial regions and the determined characteristic value.
9. A non-transitory computer-readable storage medium having stored thereon, a computer program having at least one code section for image processing, the at least one code section being executable by a computer for causing the computer to perform steps comprising:
allocating each pixel of an input image to a plurality of partial regions, wherein each of the plurality of partial regions corresponds to a predetermined range of luminance value;
determining a number of pixels of the input image, allocated to each of the plurality of partial regions, up to a decimal precision of a predetermined number of bits;
determining a characteristic value corresponding to each of the plurality of partial regions; and
performing edge-preserving smoothing on the input image by weighting and averaging characteristic values in accordance with a distance in luminance direction, using the number of pixels of the input image allocated to each of the plurality of partial regions and the determined characteristic value.