US 12,079,950 B2
Image processing method and apparatus, smart microscope, readable storage medium and device
Liang Wang, Shenzhen (CN); Jia Rui Sun, Shenzhen (CN); Yan Chun Zhu, Shenzhen (CN); and Jianhua Yao, Shenzhen (CN)
Assigned to TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED, Shenzhen (CN)
Filed by TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED, Guangdong (CN)
Filed on Mar. 3, 2022, as Appl. No. 17/685,860.
Application 17/685,860 is a continuation of application No. PCT/CN2020/027037, filed on Nov. 6, 2020.
Claims priority of application No. 202010095182.0 (CN), filed on Feb. 14, 2020.
Prior Publication US 2022/0309610 A1, Sep. 29, 2022
Int. Cl. G06T 3/40 (2024.01); G06T 7/11 (2017.01); G06V 10/44 (2022.01); G06V 10/764 (2022.01)
CPC G06T 3/40 (2013.01) [G06T 7/11 (2017.01); G06V 10/44 (2022.01); G06V 10/764 (2022.01); G06T 2207/10056 (2013.01)] 20 Claims
OG exemplary drawing
 
1. An image processing method, performed by at least one processor, the method comprising:
acquiring a digital slide comprising objects of at least two sizes, the objects comprising a first object of a first size and a second object of a second size different from the first size;
for each of the first object and the second object, acquiring images of a corresponding object in at least two scales based on the digital slide, wherein the acquired images of the corresponding object in the at least two scales comprise an image of a larger scale and an image of a smaller scale, the image of the larger scale having a smaller image size and a higher image resolution than the image of the smaller scale;
for each of the first object and the second object, determining, from the images of the corresponding object in the at least two scales, an image having an image size that corresponds to a size of the corresponding object;
obtaining a first classification result of the first object by classifying the first object in the determined image for the first object based on a classification model corresponding to a first image resolution of the determined image for the first object;
obtaining a second classification result of the second object by classifying the second object in the determined image for the second object based on a classification model corresponding to a second image resolution of the determined image for the second object; and
obtaining classification results of the first object and the second object by fusing the first classification result and the second classification result.