US 11,967,125 B2
Image processing method and system
Zhiyu Wang, Ningde (CN); Lu Li, Ningde (CN); and Jin Wei, Ningde (CN)
Assigned to CONTEMPORARY AMPEREX TECHNOLOGY CO., LIMITED, Ningde (CN)
Filed by CONTEMPORARY AMPEREX TECHNOLOGY CO., LIMITED, Ningde (CN)
Filed on Apr. 4, 2023, as Appl. No. 18/295,513.
Application 18/295,513 is a continuation of application No. PCT/CN2021/136052, filed on Dec. 7, 2021.
Prior Publication US 2023/0237763 A1, Jul. 27, 2023
Int. Cl. G06T 7/00 (2017.01); G06T 7/10 (2017.01); G06V 10/44 (2022.01); G06V 10/764 (2022.01)
CPC G06V 10/44 (2022.01) [G06T 7/10 (2017.01); G06V 10/764 (2022.01); G06T 2207/20084 (2013.01)] 10 Claims
OG exemplary drawing
 
1. An image processing method, comprising:
determining whether each pixel of an input image belongs to a target object of the input image using a segmentation algorithm, wherein the target object is composed of pixels belonging to the target object;
determining an enhanced image of the target object, wherein the enhanced image of the target object is an image in which each pixel of the target object is displayed in an enhanced manner; and
determining a positioning image of the target object by applying an integral image algorithm to the enhanced image of the target object;
wherein determining whether each pixel of the input image belongs to the target object of the input image using the segmentation algorithm comprises:
inputting the input image into a deep convolutional neural network to perform a pixel-level feature extraction on the input image; and
determining whether the pixel belongs to the target object based on a feature value of the pixel;
wherein determining the enhanced image of the target object comprises:
performing a feature extraction on the input image to determine a pixel feature graph;
performing a feature extraction on the input image to determine a context feature graph;
determining context relation information of each pixel based on the pixel feature graph and the context feature graph; and
determining the enhanced image of the target object according to the context relation information of each pixel of the input image;
wherein each pixel of the input image carries weight information, and determining the enhanced image of the target object according to the context relation information of each pixel of the input image comprises:
changing weight information of the pixels belonging to the target object, so that the target object is displayed in the enhanced manner.