US 12,114,100 B2
Image processing method in remote control, device, apparatus and program product
Qingrui Sun, Beijing (CN); Jingchao Feng, Beijing (CN); Liming Xia, Beijing (CN); and Zhuo Chen, Beijing (CN)
Assigned to Beijing Baidu Netcom Science Technology Co., Ltd., Beijing (CN)
Filed by Beijing Baidu Netcom Science Technology Co., Ltd., Beijing (CN)
Filed on Nov. 4, 2021, as Appl. No. 17/519,375.
Claims priority of application No. 202011525076.8 (CN), filed on Dec. 22, 2020.
Prior Publication US 2022/0058771 A1, Feb. 24, 2022
Int. Cl. G06K 9/00 (2022.01); G05D 1/00 (2006.01); G06T 3/4038 (2024.01); G06T 9/00 (2006.01); G06V 10/46 (2022.01); G06V 10/80 (2022.01); H04L 65/61 (2022.01); H04N 7/18 (2006.01)
CPC H04N 7/181 (2013.01) [G05D 1/0246 (2013.01); G06T 3/4038 (2013.01); G06T 9/00 (2013.01); G06V 10/462 (2022.01); G06V 10/80 (2022.01); H04L 65/61 (2022.05)] 10 Claims
OG exemplary drawing
 
1. An image processing method in remote control, performed by a cloud apparatus, comprising:
receiving image information sent by a vehicle, wherein the image information comprises multiple-channel image data collected by the vehicle;
performing a stitching process on the multiple-channel image data to obtain a stitched image; and
sending the stitched image to a remote cockpit apparatus for controlling the vehicle;
wherein the multiple-channel image data comprises a first channel image data, a second channel image data and a third channel image data, the performing the stitching process on the multiple-channel image data to obtain the stitched image comprises:
extracting a first common pixel point of the first channel image data and the second channel image data;
creating a first coordinate system based on the first common pixel point, mapping the first channel image data to the first coordinate system to obtain first image mapping data, and mapping the second channel image data to the first coordinate system to obtain second image mapping data;
replicating the first image mapping data and the second image mapping data into a first blank image, respectively, and performing weighted fusion of overlapping image data of the first image mapping data and the second image mapping data in the first blank image to obtain a first stitched sub-image;
extracting a second common pixel point of the third channel image data and the second channel image data;
creating a second coordinate system based on the second common pixel point, mapping the third channel image data to the second coordinate system to obtain third image mapping data, and mapping the second channel image data to the second coordinate system to obtain fourth image mapping data;
replicating the third image mapping data and the fourth image mapping data into a second blank image, respectively, and performing weighted fusion of overlapping image data of the third image mapping data and the fourth image mapping data in the second blank image to obtain a second stitched sub-image; and
stitching the first stitched sub-image and the second stitched sub-image to obtain the stitched image.