# US 8,542,919 B2 · Method and system for correcting lens shading > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 8,542,919 B2 - Document ID: US8542919B2 - Status: Granted - Kind: B2 - Application number: 13/163,859 - Filed: 2011-06-20 - Published: 2011-12-22 - Granted: 2013-09-24 - Earliest claimed: 2010-06-22 - Assignees: SK hynix Inc. - Applicants: Kang Hoon Seo; Min Seok Kim - Inventors: Kang Hoon Seo; Min Seok Kim - Web page: https://patentgrep.com/patent/US8542919B2 ## CPC - G06T 5/94: based on local image properties, e.g. for local contrast enhancement - G06T 2207/20021: Dividing image into blocks, subimages or windows - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL ## Abstract There is provided an image sensor according to an aspect of the present invention that includes: a reference image luminance extractor partitioning a reference image into a plurality of reference image blocks and extracting luminance values of edge pixels at the edges of each of the plurality of reference image blocks to calculate a luminance weight value for each of the edge pixels; a luminance weight value calculator calculating a luminance weight value of each pixel of the corresponding reference image block on the basis of luminance weight values of edge pixels of the corresponding reference image block with respect to each of the plurality of reference image blocks; and a luminance corrector correcting the luminance of an input image by multiplying the luminance weight value for each pixel calculated by the luminance weight value calculator with respect to a luminance value of each pixel of the input image. ## Claims ### Claim 1 An image sensor, comprising: a reference image luminance extractor, operative to partition a reference image into a plurality of reference image blocks, extract luminance values of four edge pixels of each of the plurality of reference image blocks and calculate a luminance weight value for each of the edge pixels; a luminance weight value calculator operatively coupled to the reference image luminance extractor, operative to calculate a luminance weight value of each pixel of a corresponding reference image block, on the basis of luminance weight values of four edge pixels of the corresponding reference image block, for each of the plurality of reference image blocks; and a luminance corrector operatively coupled to the luminance weight value calculator, operative to correct the luminance of an input image by multiplying the luminance weight value for each pixel, calculated by the luminance weight value calculator, with a luminance value of each pixel of the input image. ### Claim 2 (depends on 1) The image sensor of claim 1, wherein the reference image luminance extractor is operative to calculate the luminance weight value on the basis of the brightest luminance value among luminance values of all the edge pixels extracted from the reference image. ### Claim 3 (depends on 1) The image sensor of claim 1, wherein the luminance weight value calculator is operative to calculate the luminance weight values of each pixel of the corresponding reference image block, in proportion to a distance between each pixel and the four edge pixels of the corresponding reference block for each of the plurality of reference image blocks. ### Claim 4 (depends on 1) The image sensor of claim 1, wherein the luminance corrector is operative to clamp a corrected pixel value to a predetermined set minimum value when the corrected pixel value is smaller than the predetermined set minimum value. ### Claim 5 (depends on 1) The image sensor of claim 1, wherein the luminance corrector is operative to clamp a corrected pixel value to a predetermined set maximum value when the corrected pixel value is larger than the predetermined set maximum value. ### Claim 6 A method for correcting lens shading of an image sensor, comprising: partitioning a reference image into a plurality of reference image blocks and extracting luminance values of four edge pixels at the edges of each of the plurality of reference image blocks and calculating a luminance weight value for each of the edge pixels; calculating a luminance weight value of each pixel of a corresponding reference image block, on the basis of luminance weight values of four edge pixels of the corresponding reference image block, for each of the plurality of reference image blocks; and correcting the luminance of an input image by multiplying the luminance weight value for each pixel, calculated by the luminance weight value calculator, with a luminance value of each pixel of the input image, wherein at least one of the foregoing steps is performed by a processor or hardwired logic. ### Claim 7 (depends on 6) The method of claim 6, wherein calculating the luminance weight value for each of the edge pixels, comprises: calculating the luminance weight value on the basis of the brightest luminance value among luminance values of all the edge pixels extracted from the reference image. ### Claim 8 (depends on 6) The method of claim 6, wherein calculating the luminance weight value of each pixel of the corresponding reference image block, comprises: calculating the luminance weight value in proportion to a distance between each pixel and the four edge pixels of the corresponding reference block for each of the plurality of reference image blocks. ### Claim 9 (depends on 6) The method of claim 6, wherein correcting the luminance, comprises: clamping a corrected pixel value to a predetermined set minimum value when the corrected pixel value is smaller than the predetermined set minimum value. ### Claim 10 (depends on 6) The method of claim 6, wherein correcting the luminance, comprises: clamping a corrected pixel value to a predetermined maximum value when the corrected pixel value is larger than the predetermined set maximum value. ### Claim 11 An image sensor, comprising: a reference image luminance extractor, operative to partition a reference image into a plurality of reference image blocks, extract luminance values of four edge pixels at the edges of each of the plurality of reference image blocks, and calculate a luminance weight value for each of a plurality of edge pixels; a luminance weight value difference storage, operatively coupled to the reference image luminance extractor, operative to store a luminance weight value difference between a luminance weight value of an edge pixel and a luminance weight value of an adjacent edge pixel, adjacent to the edge pixel, for each of the plurality of edge pixels; a luminance weight value restorer, operatively coupled to the luminance weight value difference storage, operative to restore the luminance weight value difference to the luminance weight value; a luminance weight value calculator, operatively coupled to the luminance weight value restorer, operative to calculate a luminance weight value of each pixel of a corresponding reference image block on the basis of luminance weight values of four edge pixels of the corresponding reference image block for the plurality of reference image blocks by using the luminance weight value restored in the luminance weight value restorer; and a luminance corrector, operatively coupled to the luminance weight value calculator, operative to correct the luminance of an input image by multiplying a luminance weight value for each pixel, calculated in the luminance weight value calculator, by a luminance value of each pixel of the input image. ### Claim 12 (depends on 11) The image sensor of claim 11, wherein the size of data representing the luminance weight value difference is smaller than the size of data representing the luminance weight value. ### Claim 13 (depends on 11) The image sensor of claim 11, wherein the luminance weight value difference storage is operative to calculate and store the luminance weight value difference in accordance with a predetermined set order from a reference edge pixel to another edge pixel by using any one pixel of the plurality of edge pixels as the reference edge pixel. ### Claim 14 (depends on 13) The image sensor of claim 13, wherein the luminance weight value difference storage is further operative to store a difference between the luminance weight value of the edge pixel and a luminance weight value just prior to the corresponding edge pixel in the predetermined set order with respect to each of the plurality of edge pixels. ### Claim 15 (depends on 11) The image sensor of claim 11, wherein the reference image luminance extractor is operative to calculate the luminance weight value on the basis of the brightest luminance value among luminance values of all the edge pixels extracted from the reference image. ### Claim 16 (depends on 11) The image sensor of claim 11, wherein the luminance weight value calculator is operative to calculate the luminance weight values of four edge pixels of the corresponding reference image block in proportion to a distance between each pixel and four edge pixels with respect to each of the plurality of reference image blocks. ### Claim 17 (depends on 11) The image sensor of claim 11, wherein the luminance corrector is operative to clamp a corrected pixel value to a predetermined set minimum value when the corrected pixel value is smaller than the predetermined set minimum value. ### Claim 18 (depends on 11) The image sensor of claim 11, wherein the luminance corrector is operative to clamp the corrected pixel value to a predetermined set maximum value when the corrected pixel value is larger than the predetermined set maximum value. ### Claim 19 A method for correcting lens shading of an image sensor, comprising: partitioning a reference image into a plurality of reference image blocks, extracting luminance values of four edge pixels at the edges of each of the plurality of reference image blocks and calculating a luminance weight value for each of the edge pixels; storing a luminance weight value difference between a luminance weight value of an edge pixel and a luminance weight value of an adjacent edge pixel, adjacent to the edge pixel, for each of the plurality of edge pixels; restoring the luminance weight value difference to the luminance weight value; calculating a luminance weight value of each pixel of a corresponding reference image block on the basis of luminance weight values of four edge pixels of the corresponding reference image block for the plurality of reference image blocks by using the luminance weight value restored in the luminance weight value restorer; and correcting the luminance of an input image by multiplying a luminance weight value for each pixel, calculated in the luminance weight value calculator, by a luminance value of each pixel of the input image, wherein at least one of the foregoing steps is performed by a processor or hardwired logic. ### Claim 20 (depends on 19) The method of claim 19, wherein the size of data representing the luminance weight value difference is smaller than the size of data representing the luminance weight value. ### Claim 21 (depends on 19) The method of claim 19, wherein storing the luminance weight value difference, comprises: calculating and storing the luminance weight value difference in accordance with a predetermined set order from a reference edge pixel to another edge pixel by using any one pixel of the plurality of edge pixels as the reference edge pixel. ### Claim 22 (depends on 20) The method of claim 20, wherein storing the luminance weight value difference, comprises storing a difference between the luminance weight value of the corresponding pixel and a luminance weight value just prior to the corresponding edge pixel in the predetermined set order with respect to each of the plurality of edge pixels. ### Claim 23 (depends on 19) The method of claim 19, wherein calculating the luminance weight value for each of the edge pixels, comprises: calculating the luminance weight value on the basis of the brightest luminance value among luminance values of all the edge pixels extracted from the reference image. ### Claim 24 (depends on 19) The method of claim 19, wherein calculating the luminance weight value of each pixel of the corresponding reference image block, comprises: calculating the luminance weight value in proportion to a distance between each pixel and the four edge pixels of the corresponding reference block for each of the plurality of reference image blocks. ### Claim 25 (depends on 19) The method of claim 19, wherein correcting the luminance, comprises: clamping a corrected pixel value to a predetermined set minimum when the corrected pixel value is smaller than the predetermined set minimum value. ### Claim 26 (depends on 19) The method of claim 19, wherein correcting the luminance, comprises: clamping a corrected pixel value to a predetermined maximum value when the corrected pixel value is larger than the predetermined set maximum value.