# US 12,602,742 B2 · Image processing apparatus, binarization method, and non-transitory recording medium > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,602,742 B2 - Document ID: US12602742B2 - Status: Granted - Kind: B2 - Application number: 18/413,383 - Filed: 2024-01-16 - Published: 2024-08-15 - Granted: 2026-04-14 - Earliest claimed: 2023-02-14 - Assignees: Ricoh Company, Ltd. - Applicants: Noriko Miyagi - Inventors: Noriko Miyagi - Web page: https://patentgrep.com/patent/US12602742B2 ## CPC - G06T 5/40: using histogram techniques - G06T 5/92: based on global image properties - G06T 7/0002: {Inspection of images, e.g. flaw detection} - G06T 2207/10008: from scanner, fax or copier - G06T 2207/30176: Document - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL ## Abstract An image processing apparatus includes circuitry that calculates, based on pixel values of an input image having M gradations, an N−1 first threshold value, wherein M>N>2, compares the pixel values of the input image with the N−1 first threshold value to generate an N-valued image from the input image, and generates, based on pixel values of each of local regions of the input image and gradation values of each of local regions of the N-valued image, a binarized image obtained by binarizing the input image. ## Claims ### Claim 1 An image processing apparatus comprising: circuitry configured to: calculate, based on pixel values of an input image having M gradations, an N−1 first threshold value, wherein M>N>2; compare the pixel values of the input image with the N−1 first threshold value to generate an N-valued image from the input image; select, based on pixel values of each of local regions of the input image and gradation values of each of local regions of the N-valued image, a binarization method for a pixel of interest of the input image, and generate, based on the pixel values of each of local regions of the input image and the gradation values of each of local regions of the N-valued image, a binarized image obtained by using the selected binarization method that binarizes the pixel of interest of the input image based on pixel values of a local region including the pixel of interest in the input image and the gradation values of the local region including the pixel of interest in the N-valued image. ### Claim 2 (depends on 1) The image processing apparatus according to claim 1, wherein when a first condition is satisfied, the first condition being that a count result of the number of pixels having a gradation value equal to a gradation value of the pixel of interest located at a center of a local region in the N-valued image is smaller than or equal to a second threshold value and that the number of types of gradation values in the local region is two, the circuitry is configured to binarize the pixel of interest in the N-valued image in accordance with another gradation value different from the gradation value of the pixel of interest. ### Claim 3 (depends on 2) The image processing apparatus according to claim 2, wherein when a second condition is satisfied, the second condition being that the gradation value of the pixel of interest in the N-valued image is a first predetermined value and that the input image includes an edge having a predetermined intensity or more, the circuitry is configured to binarize the input image by adaptive binarization. ### Claim 4 (depends on 2) The image processing apparatus according to claim 2, wherein when a third condition is satisfied, the third condition being that the gradation value of the pixel of interest in the N-valued image is a second predetermined value, the circuitry is configured to binarize the input image by adaptive binarization. ### Claim 5 (depends on 1) The image processing apparatus according to claim 1, wherein when none of a first condition, a second condition, and a third condition is satisfied, the first condition being that a count result of the number of pixels having a gradation value equal to a gradation value of the pixel of interest located at a center of a local region in the N-valued image is smaller than or equal to a second threshold value and that the number of types of gradation values in the local region is two, the second condition being that the gradation value of the pixel of interest in the N-valued image is a first predetermined value and that the input image includes an edge having a predetermined intensity or more, the third condition being that the gradation value of the pixel of interest in the N-valued image is a second predetermined value, the circuitry is configured to binarize the pixel of interest in the N-valued image into a value uniquely determined with respect to a gradation value in the local region. ### Claim 6 (depends on 2) The image processing apparatus according to claim 2, wherein when a second condition is not satisfied, the second condition being that the gradation value of the pixel of interest in the N-valued image is a first predetermined value and that the input image includes an edge having a predetermined intensity or more, the circuitry is configured to determine that a change in density in the local region of the input image is more gradual than a predetermined change, and when a third condition is not satisfied, the third condition being that the gradation value of the pixel of interest in the N-valued image is a second predetermined value, the circuitry is configured to binarize the pixel of interest in the N-valued image into a value uniquely determined with respect to a gradation value in the local region. ### Claim 7 (depends on 2) The image processing apparatus according to claim 2, wherein when a second condition is satisfied, the second condition being that the gradation value of the pixel of interest in the N-valued image is a first predetermined value and that the input image includes an edge having a predetermined intensity or more, the circuitry is configured to: determine that an edge is present in the local region, and perform adaptive binarization on the local region of the input image to binarize the input image. ### Claim 8 (depends on 1) The image processing apparatus according to claim 1, wherein the circuitry is configured to create a histogram of gradation values and detect a value close to a minimum value of the histogram as the first threshold value. ### Claim 9 (depends on 8) The image processing apparatus according to claim 8, wherein the circuitry is configured to apply a discriminant analysis method while changing a reference range of the histogram to calculate a plurality of first threshold values including the N−1 first threshold value. ### Claim 10 (depends on 1) The image processing apparatus according to claim 1, wherein when a first condition is satisfied, the first condition being that a count result of the number of pixels having a gradation value equal to a gradation value of the pixel of interest located at a center of a local region in the N-valued image is smaller than or equal to a second threshold value and that the number of types of gradation values in the local region is two, the circuitry is configured to binarize the pixel of interest in the N-valued image in accordance with another gradation value different from the gradation value of the pixel of interest, when the first condition is not satisfied but a second condition is satisfied, the second condition being that the gradation value of the pixel of interest in the N-valued image is a first predetermined value and that an absolute value of a difference between an average value and a minimum value of a local region including a pixel of interest at a center in the input image is greater than or equal to a third threshold value, the circuitry is configured to binarize the input image by adaptive binarization, when the second condition is not satisfied but a third condition is satisfied, the third condition being that the gradation value of the pixel of interest in the N-valued image is a second predetermined value, the circuitry is configured to binarize the input image by adaptive binarization, and when none of the first condition, the second condition, and the third condition is satisfied, the circuitry is configured to binarize the pixel of interest in the N-valued image into a value uniquely determined with respect to a gradation value in the local region. ### Claim 11 (depends on 1) The image processing apparatus according to claim 1, wherein when a predetermined condition relating to the gradation values of pixels in the local region including the pixel of interest in the N-valued image is satisfied, the circuitry is configured to binarize the pixel of interest in the N-valued image in accordance with another gradation value different from the gradation value of the pixel of interest. ### Claim 12 A binarization method performed by an image processing apparatus, the binarization method comprising: calculating, based on pixel values of an input image having M gradations, an N−1 first threshold value, wherein M>N>2; comparing the pixel values of the input image with the N−1 first threshold value to generate an N-valued image from the input image; selecting, based on pixel values of each of local regions of the input image and gradation values of each of local regions of the N-valued image, a binarization method for a pixel of interest of the input image, and generating, based on the pixel values of each of local regions of the input image and the gradation values of each of local regions of the N-valued image, a binarized image obtained by using the selected binarization method that binarizes the pixel of interest of the input image based on pixel values of the local region including the pixel of interest in the input image and the gradation values of the local region including the pixel of interest in the N-valued image. ### Claim 13 A non-transitory recording medium storing a plurality of instructions which, when executed by one or more processors, causes the one or more processors to perform a binarization method comprising: calculating, based on pixel values of an input image having M gradations, an N−1 first threshold value, wherein M>N>2; comparing the pixel values of the input image with the N−1 first threshold value to generate an N-valued image from the input image; selecting, based on pixel values of each of local regions of the input image and gradation values of each of local regions of the N-valued image, a binarization method for a pixel of interest of the input image, and generating, based on the pixel values of each of local regions of the input image and the gradation values of each of local regions of the N-valued image, a binarized image obtained by using the selected binarization method that binarizes the pixel of interest of the input image based on pixel values of the local region including the pixel of interest in the input image and the gradation values of the local region including the pixel of interest in the N-valued image.