US 12,327,332 B2
Multi-modality image processing method and device
Yao Zhang, Beijing (CN); and Jiang Tian, Beijing (CN)
Assigned to LENOVO (BEIJING) LIMITED, Beijing (CN)
Filed by Lenovo (Beijing) Limited, Beijing (CN)
Filed on Mar. 3, 2022, as Appl. No. 17/686,395.
Claims priority of application No. 202111154668.8 (CN), filed on Sep. 29, 2021.
Prior Publication US 2023/0103262 A1, Mar. 30, 2023
Int. Cl. G06T 5/50 (2006.01); G06F 18/214 (2023.01); G06F 18/22 (2023.01); G06F 18/25 (2023.01); G06V 10/40 (2022.01)
CPC G06T 5/50 (2013.01) [G06F 18/214 (2023.01); G06F 18/22 (2023.01); G06F 18/251 (2023.01); G06F 18/253 (2023.01); G06V 10/40 (2022.01); G06T 2207/10081 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20221 (2013.01)] 18 Claims
OG exemplary drawing
 
1. An image processing method comprising:
obtaining a first quantity of to-be-analyzed images, each to-be-analyzed image corresponding to a different target modality of a target imaging object;
performing fusion and enhancement processing on the first quantity of to-be-analyzed images through an image analysis model to obtain a first target image, the first target image being used to enhance display of a distribution area of an analysis object of the first quantity of to-be-analyzed images, the analysis object belonging to the target imaging object, the image analysis model being obtained by training a second quantity of sample images corresponding to different sample modalities, the first quantity being less than or equal to the second quantity, and the target modality belonging to the sample modalities, including:
in response to the first quantity being greater than or equal to 2 and less than or equal to the second quantity, processing the corresponding to-be-analyzed images using feature extraction sub-models to obtain a first quantity of reference images, the feature extraction sub-models corresponding to the first quantity of to-be-analyzed images of different target modalities in the image analysis model, a feature extraction sub-model corresponding to each to-be-analyzed image representing an association relationship between a second quantity of to-be-analyzed images of different sample modalities; and
processing the first quantity of reference images through a modal perception sub-model of the image analysis model to obtain the first target image.