US 12,462,569 B2
Image processing method, electronic device, and computer program product
Bin He, Shanghai (CN); Jiacheng Ni, Shanghai (CN); Wenlei Wu, Shanghai (CN); and Zhen Jia, Shanghai (CN)
Assigned to Dell Products L.P., Round Rock, TX (US)
Filed by Dell Products L.P., Round Rock, TX (US)
Filed on Apr. 3, 2023, as Appl. No. 18/129,938.
Claims priority of application No. 202310207543.X (CN), filed on Feb. 27, 2023.
Prior Publication US 2024/0290104 A1, Aug. 29, 2024
Int. Cl. G06V 10/778 (2022.01); G06V 10/764 (2022.01); G06V 20/54 (2022.01); G08G 1/16 (2006.01)
CPC G06V 20/54 (2022.01) [G06V 10/764 (2022.01); G06V 10/778 (2022.01); G08G 1/16 (2013.01)] 20 Claims
OG exemplary drawing
 
1. An image processing method, comprising:
receiving a sub-image containing a target object and location information of the target object from a first computing node in a first layer of a multi-layer computing architecture, the sub-image being intercepted from a monitoring image by the first computing node through target detection, the monitoring image being acquired by a roadside device and containing the target object, and the first computing node being located near the roadside device, the target detection being performed using a first machine learning model deployed at the first computing node in the first layer of the multi-layer computing architecture and configured for object detection;
determining classification information of the target object based on the sub-image at a second computing node different from the first computing node, the second computing node being in a second layer of the multi-layer computing architecture, the determining of the classification information of the target object being performed using a second machine learning model, different than the first machine learning model, deployed at the second computing node in the second layer of the multi-layer computing architecture and configured for object classification; and
generating safety hint information for the target object at the second computing node at least based on the classification information and the location information;
wherein the first and second machine learning models are trained in at least one additional computing node in at least one additional layer of the multi-layer computing architecture.