Granted Patent B2

US 12,488,563 · App. 17/897,121 · Granted Dec 2, 2025

Inventors: Zuo Xu (Qinhuangdao, CN); Ning Wang (Qinhuangdao, CN); Tao Tian (Qinhuangdao, CN); Xi Li (Qinhuangdao, CN); Lihua Chen (Qinhuangdao, CN)

Assignee: CITIC Dicastal CO., LTD

G06V10/761G06F16/532G06T7/62G06V10/46G06T2207/30242

Abstract

Claims (42)

1 . A hub image retrieval method, comprising:

performing feature extraction on hub image to be processed containing a target hub, and obtaining a hub feature to be processed containing N sub-hub features, wherein the N sub-hub features at least include information characterizing the features of a hub window of the target hub, N is an integer greater than zero;

determining similarities between candidate hub features of each candidate hub image in a plurality of candidate hub images and the hub features to be processed;

selecting a candidate hub image matching the hub image to be processed from the plurality of candidate hub images based on the determined respective similarities;

wherein the sub-hub feature comprises a second sub-hub feature characterizing a shape of a hub window of the target hub in the hub image to be processed;

wherein when the sub-hub feature comprises the second sub-hub feature, in the process of performing feature extraction on a hub image to be processed containing a target hub and obtaining a hub feature to be processed containing N sub-hub features comprises:

extracting a first hash of each hub window of the target hub in the hub image to be processed;

determining a hub window array of the target hub based on each extracted first hash; wherein the determined hub window array comprises a single window array or a multi-window array; and

determining a second hash of the hub window array as the second sub-hub feature.

2 . The method of claim 1 , wherein the sub-hub feature further comprises:

a first sub-hub feature characterizing a number of hub windows of the target hub in the hub image to be processed.

3 . The method of claim 1 , wherein the sub-hub feature further comprises at least one of:

a third sub-hub feature characterizing hub size information of the target hub in the hub image to be processed;

a fourth sub-hub feature characterizing information of the number of bolt holes in the target hub in the hub image to be processed.

4 . The method of claim 3 , wherein when the hub feature to be processed comprises the third sub-hub feature,

the performing feature extraction on a hub image to be processed containing a target hub, and obtaining a hub feature to be processed further comprises:

determining a center point of the target hub in the image to be processed;

extracting an outer radius and an inner radius of the target hub based on the center point;

determining the third sub-hub feature based on the outer radius and the inner radius.

5 . The method of claim 1 , wherein determining similarities between candidate hub features of each candidate hub image and the hub features to be processed comprises:

respectively taking each candidate hub feature as a current candidate hub feature and executing a similarity determination operation; wherein, the similarity determination operation includes:

respectively carrying out N target operations on the current candidate hub features, and determining N sub-similarities corresponding to the current candidate hub features; a target operation includes: determining a sub-hub feature of the N sub-hub features as a target sub-hub feature, and determining the similarity between the target sub-hub feature contained in the current candidate hub feature and the target sub-hub feature contained in the hub feature to be processed as a sub-similarity corresponding to the current candidate hub feature; wherein, the features of the target sub-hubs in different target operations are different;

determining a similarity between the current candidate hub features and the hub features to be processed based on N sub-similarities corresponding to the current candidate hub features.

6 . The method according to claim 5 , wherein determining a similarity between the current candidate hub features and the hub features to be processed based on N sub-similarities corresponding to the current candidate hub features comprises:

carrying out weighted summation on the N sub-similarities according to the respective corresponding weights of the N sub-hub features to obtain a similarity between the current candidate hub features and the hub features to be processed; or

determining the number of sub-similarities larger than a first similarity threshold in the N sub-similarities as the similarity between the current candidate hub feature and the hub feature to be processed.

7 . The method of claim 1 , wherein selecting a candidate hub image matching the hub image to be processed from the plurality of candidate hub images based on the determined respective similarities comprises:

determining a similarity greater than a second similarity threshold in each similarity as a target similarity, or determining a maximum similarity degree in each similarity as the target similarity;

determining the candidate hub image corresponding to the target similarity as a candidate hub image matching the hub image to be processed.

8 . The method of claim 1 , wherein the N is an integer greater than 1, and the priority of different sub-hub features in the N sub-hub features is different, and before determining similarities between candidate hub features of each candidate hub image in a plurality of candidate hub images and the hub features to be processed, further comprises:

selecting at least one of N sub-hub features included in the hub feature to be processed as a basic sub-hub feature based on the priority of different sub-hub features;

screening out a part of the candidate hub images from the plurality of candidate hub images based on the basic sub-hub feature;

the determining similarities between candidate hub features of each candidate hub image in a plurality of candidate hub images and the hub features to be processed comprises:

determining a similarity between a candidate hub feature of each candidate hub image in the plurality of candidate hub images and the hub feature to be processed.

9 . A computer device, comprising a memory, a processor, and a computer program stored in the memory and operable on the processor, wherein the processor implements the steps of the method according to claim 1 when executing the program.

10 . A computer device, comprising a memory, a processor, and a computer program stored in the memory and operable on the processor, wherein the processor implements the steps of the method according to claim 2 when executing the program.

11 . A computer device, comprising a memory, a processor, and a computer program stored in the memory and operable on the processor, wherein the processor implements the steps of the method according to claim 3 when executing the program.

12 . A computer device, comprising a memory, a processor, and a computer program stored in the memory and operable on the processor, wherein the processor implements the steps of the method according to claim 4 when executing the program.

13 . A computer device, comprising a memory, a processor, and a computer program stored in the memory and operable on the processor, wherein the processor implements the steps of the method according to claim 5 when executing the program.

14 . A computer device, comprising a memory, a processor, and a computer program stored in the memory and operable on the processor, wherein the processor implements the steps of the method according to claim 6 when executing the program.

15 . A computer device, comprising a memory, a processor, and a computer program stored in the memory and operable on the processor, wherein the processor implements the steps of the method according to claim 7 when executing the program.

16 . A computer device, comprising a memory, a processor, and a computer program stored in the memory and operable on the processor, wherein the processor implements the steps of the method according to claim 8 when executing the program.

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (13)