# US 12,223,619 B2 · Image processing apparatus, image processing method, and display apparatus based on the same > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,223,619 B2 - Document ID: US12223619B2 - Status: Granted - Kind: B2 - Application number: 17/888,366 - Filed: 2022-08-15 - Published: 2023-06-08 - Granted: 2025-02-11 - Earliest claimed: 2021-12-08 - Assignees: LG Display Co., Ltd. - Applicants: LG Display Co., Ltd. - Inventors: Seong Gyun Kim - Web page: https://patentgrep.com/patent/US12223619B2 ## CPC - G06T 3/4076: using the original low-resolution images to iteratively correct the high-resolution images - G06T 3/4007: based on interpolation, e.g. bilinear interpolation (image demosaicing G06T3/4015; edge-driven or edge-based scaling G06T3/403) - G06T 3/4046: using neural networks - G06T 3/4053: based on super-resolution, i.e. the output image resolution being higher than the sensor resolution - G06T 5/50: using two or more images, e.g. averaging or subtraction - G06T 5/60: using machine learning, e.g. neural networks - G06T 5/70: Denoising; Smoothing - G06T 2207/20081: Training; Learning - G06T 2207/20084: Artificial neural networks [ANN] - G06T 2207/20224: Image subtraction - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL ## Abstract The present disclosure provides an image processing apparatus, and image processing method, and a display apparatus. The image processing apparatus includes a first image processor up-sampling an original low-resolution image on the basis of deep learning-based learning data to generate a first high-resolution image, a second image processor interpolating the original low-resolution image to generate a second high-resolution image, a third image processor generating a difference image between the first high-resolution image and the second high-resolution image, extracting a high frequency component from the difference image, and amplifying the extracted high frequency component, and a fourth image processor adding the amplified high frequency component to the first high-resolution image to generate a target high-resolution image. ## Claims ### Claim 1 An image processing apparatus comprising: a first image processor configured to up-sample an original low-resolution image on the basis of deep learning-based learning data to generate a first high-resolution image; a second image processor configured to interpolate the original low-resolution image to generate a second high-resolution image; a third image processor configured to generate a difference image between the first high-resolution image and the second high-resolution image, extract a high frequency component from the difference image, and amplify the extracted high frequency component; and a fourth image processor configured to add the amplified high frequency component to the first high-resolution image to generate a target high-resolution image. ### Claim 2 (depends on 1) The image processing apparatus of claim 1, wherein the third image processor multiplies the extracted high frequency component by a predetermined gain value to generate the amplified high frequency component. ### Claim 3 (depends on 1) The image processing apparatus of claim 1, wherein the learning data comprises an image pair including a learning low-resolution image and a learning high-resolution image. ### Claim 4 (depends on 1) The image processing apparatus of claim 1, wherein the third image processor further comprises a coring circuit removing noise concentrating near a threshold value of the extracted high frequency component on the basis of a predetermined coring algorithm before the extracted high frequency component is amplified. ### Claim 5 An image processing apparatus comprising: a first image processor configured to generate a denoising low-resolution image where noise is removed from an original low-resolution image; a second image processor configured to up-sample the denoising low-resolution image on the basis of deep learning-based learning data to generate a first high-resolution image; a third image processor configured to up-sample the original low-resolution image on the basis of the learning data to generate a second high-resolution image; a fourth image processor configured to weighted-average the first high-resolution image and the second high-resolution image to generate a third high-resolution image; a fifth image processor configured to interpolate the denoising low-resolution image to generate a fourth high-resolution image; a sixth image processor configured to generate a difference image between the first high-resolution image and the fourth high-resolution image, extract a high frequency component from the difference image, and amplify the extracted high frequency component; and a seventh image processor configured to add the amplified high frequency component to the third high-resolution image to generate a target high-resolution image. ### Claim 6 (depends on 5) The image processing apparatus of claim 5, wherein the fourth image processor multiplies the first high-resolution image by a first weight to calculate a first weighting result, multiplies the second high-resolution image by a second weight to calculate a second weighting result, and averages the first weighting result and the second weighting result to generate the third high-resolution image, and a sum of the first weight and the second weight is 1. ### Claim 7 (depends on 5) The image processing apparatus of claim 5, wherein the sixth image processor multiplies the extracted high frequency component by a predetermined gain value to generate the amplified high frequency component. ### Claim 8 (depends on 5) The image processing apparatus of claim 5, wherein the sixth image processor further comprises a coring circuit removing noise concentrating near a threshold value of the extracted high frequency component on the basis of a predetermined coring algorithm before the extracted high frequency component is amplified. ### Claim 9 An image processing method comprising: up-sampling an original low-resolution image on the basis of deep learning-based learning data to generate a first high-resolution image; interpolating the original low-resolution image to generate a second high-resolution image; generating a difference image between the first high-resolution image and the second high-resolution image, extracting a high frequency component from the difference image, and amplifying the extracted high frequency component; and adding the amplified high frequency component to the first high-resolution image to generate a target high-resolution image. ### Claim 10 (depends on 9) The image processing method of claim 9, wherein the amplifying the extracted high frequency component comprises multiplying the extracted high frequency component by a predetermined gain value. ### Claim 11 (depends on 9) The image processing method of claim 9, further comprising removing noise concentrating near a threshold value of the extracted high frequency component on the basis of a predetermined coring algorithm before the extracted high frequency component is amplified. ### Claim 12 An image processing method comprising: generating a denoising low-resolution image where noise is removed from an original low-resolution image; up-sampling the denoising low-resolution image on the basis of deep learning-based learning data to generate a first high-resolution image; up-sampling the original low-resolution image on the basis of the learning data to generate a second high-resolution image; weighted-averaging the first high-resolution image and the second high-resolution image to generate a third high-resolution image; interpolating the denoising low-resolution image to generate a fourth high-resolution image; generating a difference image between the first high-resolution image and the fourth high-resolution image, extracting a high frequency component from the difference image, and amplifying the extracted high frequency component; and adding the amplified high frequency component to the third high-resolution image to generate a target high-resolution image. ### Claim 13 (depends on 12) The image processing method of claim 12, wherein the weighted-averaging the first high-resolution image and the second high-resolution image comprises: multiplying the first high-resolution image by a first weight to calculate a first weighting result; multiplying the second high-resolution image by a second weight to calculate a second weighting result; and averaging the first weighting result and the second weighting result to generate the third high-resolution image, and a sum of the first weight and the second weight is 1. ### Claim 14 (depends on 12) The image processing method of claim 12, wherein the amplifying the extracted high frequency component comprises multiplying the extracted high frequency component by a predetermined gain value. ### Claim 15 (depends on 12) The image processing method of claim 12, further comprising removing noise concentrating near a threshold value of the extracted high frequency component on the basis of a predetermined coring algorithm before the extracted high frequency component is amplified. ### Claim 16 (depends on 1) A display apparatus comprising the image processing apparatus of claim 1. ### Claim 17 (depends on 5) A display apparatus comprising the image processing apparatus of claim 5.