# US 12,118,696 B2 · Image signal processor, method for operating image signal processor and image sensing device > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,118,696 B2 - Document ID: US12118696B2 - Status: Granted - Kind: B2 - Application number: 17/827,101 - Filed: 2022-05-27 - Published: 2023-06-01 - Granted: 2024-10-15 - Earliest claimed: 2021-11-26 - Assignees: Samsung Electronics Co., Ltd. - Applicants: Samsung Electronics Co., Ltd. - Inventors: So Jin Moon; Hyoung Seok Ko - Web page: https://patentgrep.com/patent/US12118696B2 ## CPC - G06T 5/70: Denoising; Smoothing - H04N 23/84: for processing colour signals - G06T 5/73: Deblurring; Sharpening - G06T 5/92: based on global image properties - H04N 9/646: {for image enhancement, e.g. vertical detail restoration, cross-colour elimination, contour correction, chrominance trapping filters} - H04N 23/82: for controlling camera response irrespective of the scene brightness, e.g. gamma correction - H04N 25/60: Noise processing, e.g. detecting, correcting, reducing or removing noise - G06T 5/20: using local operators - G06T 2207/20192: Edge enhancement; Edge preservation - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION ## Abstract An image signal processor includes a gradation detection module configured to receive a pixel signal from an external device and a correction module configured to receive the pixel signal. The correction module is connected to the gradation detection module. The gradation detection module is configured to estimate a gradient of the pixel signal, correct the pixel signal based on the gradient to generate a corrected pixel signal, calculate a first variance value based on the pixel signal, calculate a second variance value based on the corrected pixel signal, calculate a gradation probability value based on a comparison result between the first variance value and the second variance value, and generate a gradation map as information about the gradation probability value of the pixel signal. The correction module is configured to correct the pixel signal based on the gradation map. ## Claims ### Claim 1 An image signal processor comprising: a gradation detection module configured to receive a pixel signal from an external device; and a correction module connected to the gradation detection module and configured to receive the pixel signal, wherein: the gradation detection module is configured to: estimate a gradient of the pixel signal; correct the pixel signal based on the gradient to generate a corrected pixel signal; calculate a first variance value based on the pixel signal; calculate a second variance value based on the corrected pixel signal; calculate a gradation probability value based on a comparison result between the first variance value and the second variance value; and generate a gradation map as information about the gradation probability value of the pixel signal, and the correction module is configured to correct the pixel signal based on the gradation map. ### Claim 2 (depends on 1) The image signal processor of claim 1, wherein the gradation probability value is greater when the first variance value is greater than the second variance value than when the second variance value is greater than the first variance value. ### Claim 3 (depends on 2) The image signal processor of claim 2, wherein the gradation probability value is greater when the first variance value is greater than the second variance value than when the first variance value is equal to the second variance value. ### Claim 4 (depends on 1) The image signal processor of claim 1, wherein the correction module is configured to generate a final corrected pixel signal based on the pixel signal and the gradation probability value. ### Claim 5 (depends on 1) The image signal processor of claim 1, wherein the correction module includes: a correction filter configured to correct the pixel signal to generate a first corrected pixel signal; and a blending module configured to perform correction based on the pixel signal, the first corrected pixel signal, and the gradation map to generate a final corrected pixel signal. ### Claim 6 (depends on 5) The image signal processor of claim 5, wherein the final corrected pixel signal corresponds to a difference value between: (1) a product of the pixel signal and the gradation probability value included in the gradation map and (2) a product of the first corrected pixel signal and the gradation probability value. ### Claim 7 (depends on 1) The image signal processor of claim 1, wherein the correction module includes: a correction filter configured to correct the pixel signal to generate a first corrected pixel signal and to correct the corrected pixel signal from the gradation detection module to generate a second corrected pixel signal; and a blending module configured to perform correction based on the first corrected pixel signal, the second corrected pixel signal, and the gradation map to generate a final corrected pixel signal. ### Claim 8 (depends on 1) The image signal processor of claim 1, wherein: the pixel signal includes a first pixel value and a plurality of second pixel values, and the estimating of the gradient of the pixel signal includes deriving a sum of gradients of the second pixel values with respect to the first pixel value. ### Claim 9 (depends on 1) The image signal processor of claim 1, wherein the correction module includes: a pixel correction module configured to estimate the gradient of the pixel signal and to correct the pixel signal based on the gradient to generate a first corrected pixel signal; a correction filter configured to correct the pixel signal to generate a second corrected pixel signal and to correct the first corrected pixel signal to generate a third corrected pixel signal; and a blending module configured to perform correction based on the second corrected pixel signal, the third corrected pixel signal, and the gradation map to generate a final corrected pixel signal. ### Claim 10 An image sensing device comprising: an image sensor including different first to third areas; and an image signal processor configured to receive a first pixel signal generated from the first area, a second pixel signal generated from the second area, and a third pixel signal generated from the third area, wherein: the image signal processor is configured to: perform correction on the first pixel signal to generate a first corrected pixel signal, perform correction on the second pixel signal to generate a second corrected pixel signal, and perform correction on the third pixel signal to generate a third corrected pixel signal; calculate a first pixel signal variance value based on the first pixel signal, calculate a second pixel signal variance value based on the second pixel signal, and calculate a third pixel signal variance value based on the third pixel signal; calculate a first corrected pixel signal variance value based on the first corrected pixel signal, calculate a second corrected pixel signal variance value based on the second corrected pixel signal, and calculate a third corrected pixel signal variance value based on the third corrected pixel signal; and calculate a first gradation probability value based on a comparison result between the third pixel signal variance value and the third corrected pixel signal variance value, and the third pixel signal variance value is greater than the third corrected pixel signal variance value. ### Claim 11 (depends on 10) The image sensing device of claim 10, wherein the image signal processor is configured to: calculate a second gradation probability value based on a comparison result between the first pixel signal variance value and the first corrected pixel signal variance value; and calculate a third gradation probability value based on a comparison result between the second pixel signal variance value and the second corrected pixel signal variance value. ### Claim 12 (depends on 11) The image sensing device of claim 11, wherein a difference between the third pixel signal variance value and the third corrected pixel signal variance value is greater than each of: (1) a difference between the first pixel signal variance value and the first corrected pixel signal variance value and (2) a difference between the second pixel signal variance value and the second corrected pixel signal variance value. ### Claim 13 (depends on 12) The image sensing device of claim 12, wherein the first gradation probability value is greater than each of the second and third gradation probability values. ### Claim 14 (depends on 11) The image sensing device of claim 11, wherein the image signal processor is configured to: correct the third pixel signal based on the first gradation probability value; correct the first pixel signal based on the second gradation probability value; and correct the second pixel signal based on the third gradation probability value. ### Claim 15 (depends on 11) The image sensing device of claim 11, wherein the image signal processor is configured to: generate a gradation map including the second gradation probability value of the first area, the third gradation probability value of the second area, and the first gradation probability value of the third area; and correct the first to third pixel signals based on the gradation map. ### Claim 16 (depends on 10) The image sensing device of claim 10, wherein the image signal processor is configured to correct the third pixel signal based on the first gradation probability value. ### Claim 17 (depends on 10) The image sensing device of claim 10, wherein: the generating of the first corrected pixel signal includes estimating a first gradient of the first pixel signal and correcting the first pixel signal based on the first gradient, the generating of the second corrected pixel signal includes estimating a second gradient of the second pixel signal and correcting the second pixel signal based on the second gradient, and the generating of the third corrected pixel signal includes estimating a third gradient of the third pixel signal and correcting the third pixel signal based on the third gradient. ### Claim 18 A method for operating an image signal processor, the method comprising: receiving a first pixel signal including a plurality of first pixel values, a second pixel signal including a plurality of second pixel values, and a third pixel signal including a plurality of third pixel values; estimating a first gradient of the first pixel values, estimating a second gradient of the second pixel values, and estimating a third gradient of the third pixel values; correcting the first pixel values based on the first gradient to generate first corrected pixel values, correcting the second pixel values based on the second gradient to generate second corrected pixel values, and correcting the third pixel values based on the third gradient to generate third corrected pixel values; generating a first variance value based on the first pixel values, generating a second variance value based on the second pixel values, and generating a third variance value based on the third pixel values; generating a fourth variance value based on the first corrected pixel values, generating a fifth variance value based on the second corrected pixel values, and generating a sixth variance value based on the third corrected pixel values; determining the first pixel signal as a flat area based on a comparison result between the first variance value and the fourth variance value; determining the second pixel signal as an edge area based on a comparison result between the second variance value and the fifth variance value; and determining the third pixel signal as a gradation area based on a comparison result between the third variance value and the sixth variance value. ### Claim 19 (depends on 18) The method of claim 18, wherein: the third variance value is greater than the sixth variance value, and a difference between the third variance value and the sixth variance value is greater than each of: (1) a difference between the first variance value and the fourth variance value and (2) a difference between the second variance value and the fifth variance value. ### Claim 20 (depends on 18) The method of claim 18, wherein the method further comprises performing correction on the first pixel signal and the third pixel signal while not performing correction on the second pixel signal.