US 12,462,338 B2
Super-resolution image reconstruction method and device
Jinfeng Liu, Guangdong (CN)
Assigned to TCL China Star Optoelectronics Technology Co., Ltd., Shenzhen (CN)
Appl. No. 17/429,950
Filed by TCL China Star Optoelectronics Technology Co., Ltd., Guangdong (CN)
PCT Filed Jul. 21, 2021, PCT No. PCT/CN2021/107498
§ 371(c)(1), (2) Date Dec. 29, 2022,
PCT Pub. No. WO2022/267136, PCT Pub. Date Dec. 29, 2022.
Claims priority of application No. 202110702940.5 (CN), filed on Jun. 24, 2021.
Prior Publication US 2024/0046414 A1, Feb. 8, 2024
Int. Cl. G06T 3/4053 (2024.01); G06T 3/4023 (2024.01)
CPC G06T 3/4053 (2013.01) [G06T 3/4023 (2013.01)] 10 Claims
OG exemplary drawing
 
1. A super-resolution image reconstruction method, comprising:
S1, collecting an original image to be reconstructed, the original image comprising a plurality of pixels arranged in an array; wherein each row of pixels comprises a plurality of row pixel groups composed of continuous pixels, and each column of pixels comprises a plurality of column pixel groups composed of continuous pixels;
S2, obtaining, according to visible relationship between pixels in each of the plurality of row pixel groups, gray scale values corresponding to newly added row pixels between two adjacent pixels in each row of pixels, and determining, according to visible relationship between pixels in each of the plurality of column pixel groups, gray scale values corresponding to newly added column pixels between two adjacent pixels in each column of pixels;
S3, determining, according to the newly added row pixels and the newly added column pixels, a gray scale value corresponding to a newly added central pixel in a surrounding area composed of newly added row pixels between two adjacent pixels in two adjacent rows and newly added column pixels between two adjacent pixels in two adjacent columns; wherein the two adjacent pixels in the two adjacent rows and the two adjacent pixels in the two adjacent columns overlap with each other; and
S4, inserting the newly added row pixels between two adjacent pixels in each row of pixels, inserting the newly added column pixels between adjacent two pixels in each column of pixels, and inserting the central pixel in the surrounding area composed of the newly added row pixels and the newly added column pixels, in order to obtain a super-resolution image reconstructed corresponding to the original image;
wherein the step S2 of obtaining, according to the visible relationship between the pixels in each of the plurality of row pixel groups, the gray scale values corresponding to the newly added row pixels between the two adjacent pixels in each row of pixels, specifically comprises:
S21, determining, according to a gray scale value of each pixel and a row position of each pixel in a pixel row, at least one visible row pixel of each pixel in a row pixel group; wherein each pixel and the at least one the visible row pixel corresponding to each pixel satisfy a preset visible condition; and
S22, according to the gray scale value and the row position of each pixel, and a gray scale value and a row position of the at least one visible row pixel corresponding to each pixel, obtaining a gray scale value corresponding to a newly added row pixel of any row position between two adjacent pixels in each row of pixels;
wherein in the step S21, the preset visible condition comprises: gray scale values of pixels located between the row position of each pixel and the row position of the at least one visible row pixel corresponding to each pixel are all less than the gray scale value of each pixel and the gray scale value of each visible row pixel.