| CPC G06V 20/695 (2022.01) [A61B 90/20 (2016.02); G02B 21/365 (2013.01); G06T 7/0012 (2013.01); G06T 7/11 (2017.01); G06T 7/13 (2017.01); G06T 7/155 (2017.01); G06V 10/22 (2022.01); G06V 10/44 (2022.01); G06V 20/698 (2022.01); A61B 2576/02 (2013.01); G06T 2207/10056 (2013.01); G06T 2207/20021 (2013.01); G06T 2207/30004 (2013.01); G06T 2207/30096 (2013.01); G06T 2207/30242 (2013.01); G06V 2201/03 (2022.01)] | 18 Claims |

|
1. A method for obtaining a feature of duct tissue based on computer vision, the method comprising:
obtaining an image that includes the duct tissue;
determining at least two feature obtaining regions that are adapted to duct morphology of the duct tissue in an image region corresponding to the duct tissue in the image;
obtaining cell features of cells of the duct tissue in the at least two feature obtaining regions, respectively;
identifying image sub-regions with brightness greater than a brightness threshold as sieve-like pore regions in the image region corresponding to the duct tissue;
obtaining a feature of sieve-like pores in the duct tissue based on the image sub-regions of the image region corresponding to the duct tissue in the image;
obtaining the feature of the duct tissue based on the feature of the sieve-like pores and the cell features in the at least two feature obtaining regions;
selecting, from a plurality of duct tissue classifiers associated with different division methods of the image region, a duct tissue classifier based on which division method of the different division methods is used for determining the at least two feature obtaining regions; and
classifying the duct tissue using the selected duct tissue classifier based on the obtained feature of the duct tissue.
|