Unlock instant, AI-driven research and patent intelligence for your innovation.

A clustering method, device, electronic device, and storage medium for similar background pictures

A technology of similar pictures and clustering methods, applied in the field of clustering methods of similar pictures with backgrounds, storage media, devices and electronic equipment, can solve the problem that pictures cannot be processed well, clustering algorithms cannot cope well, and affect the final result. results, etc.

Active Publication Date: 2022-07-01
安徽深信科创信息技术有限公司
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] These clustering algorithms cannot deal with the problem of "normal pictures" on the graph background clustering well. "Normal pictures" means that there are a large number of pictures in the picture background clustering that do not belong to any cluster. Class images don't handle well
[0004] Common clustering algorithms cannot deal with the problem of "normal pictures" in graph background clustering well. "Normal pictures" are pictures that are normally used for business. Such pictures account for the vast majority of pictures and do not belong to any fraudulent clusters. , they will largely affect the final result

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A clustering method, device, electronic device, and storage medium for similar background pictures
  • A clustering method, device, electronic device, and storage medium for similar background pictures
  • A clustering method, device, electronic device, and storage medium for similar background pictures

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0047] See figure 1 , figure 1 It is a schematic flowchart of a clustering method for background similar pictures provided by an embodiment of the present invention. The embodiment of the present invention provides a clustering method for similar background pictures, and the clustering method for similar background pictures may specifically include steps 1 to 4, wherein:

[0048] Step 1. Construct an undirected graph G, wherein the undirected graph G is represented by an adjacency matrix, and the picture is a node of the undirected graph G.

[0049] Specifically, an adjacency matrix needs to be obtained first, and the adjacency matrix is ​​used to represent a graph. The graph includes nodes and edges, and the graph represents the relationship between the nodes. It is assumed that the picture is a node of the graph, and the edge is the edge of the graph. The degree of correlation between pictures, the greater the degree of correlation, the greater the similarity between pictu...

Embodiment 2

[0085] See image 3 , image 3 It is a schematic diagram of a clustering device for background similar pictures provided by an embodiment of the present invention. The clustering device for pictures with similar backgrounds includes:

[0086] a building module for constructing an undirected graph G, wherein the undirected graph G is represented by an adjacency matrix, and the picture is a node of the undirected graph G;

[0087] The removal module is used to remove all nodes whose core degree is less than k0 in the undirected graph G to obtain several subgraphs G1, wherein the subgraph G1 is a strong relationship cluster, and k0 is the relationship between affinity and frequency the turning point of the graph;

[0088] The clustering module is configured to divide the first non-strong relationship node into the corresponding sub-graph G1 according to the high confidence threshold of the affinity and frequency relationship graph, wherein the high confidence threshold is the ...

Embodiment 3

[0093] See Figure 4 , Figure 4 It is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. The electronic device 1100 includes: a processor 1101, a communication interface 1102, a memory 1103 and a communication bus 1104, wherein the processor 1101, the communication interface 1102, and the memory 1103 complete mutual communication through the communication bus 1104;

[0094] memory 1103 for storing computer programs;

[0095] The processor 1101 is configured to implement the above method steps when executing the computer program.

[0096] When the processor 1101 executes the computer program, the following steps are implemented:

[0097] Step 1, build undirected graph G, wherein, described undirected graph G is represented with adjacency matrix, and picture is the node of described undirected graph G;

[0098] Step 2. Remove all the nodes whose core degree is less than k0 in the undirected graph G to obtain several su...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a clustering method, device, electronic device, and storage medium for pictures with similar backgrounds. The clustering method includes: constructing an undirected graph G, wherein the undirected graph G is represented by an adjacency matrix, and the pictures are all The nodes of the undirected graph G; remove all the nodes whose core degree is less than k0 in the undirected graph G, and obtain several subgraphs G1, wherein the subgraphs G1 are strong relationship clusters, and k0 is the affinity and The turning point of the frequency relation graph; according to the high confidence threshold of the affinity and frequency relation graph, the first non-strong relation node is divided into the corresponding subgraph G1, wherein the high confidence threshold is the affinity The highest point of the graph of degree and frequency. The present invention will combine the strong correlation subgraph algorithm to mine the strong correlation relationship between different pictures, on the basis of the obtained adjacency matrix, use the uncorrelated subgraph algorithm to find the strong relationship of the corresponding entities, cluster the pictures and solve the problem of " normal picture" problem.

Description

technical field [0001] The invention belongs to the technical field of digital image processing, and in particular relates to a clustering method, device, electronic device, and storage medium of similar background pictures. Background technique [0002] At present, for image background clustering, more common clustering algorithms are generally used, such as k-means, k-medoids, spectral clustering, and affinity diffusion clustering algorithm (Affinity). Propagation) etc. to process feature information. [0003] These clustering algorithms cannot deal well with the "normal picture" problem in the background clustering of the picture. "Normal picture" means that there are a large number of pictures that do not belong to any cluster in the picture background clustering. Class pictures don't work well. [0004] Common clustering algorithms cannot deal with the problem of "normal pictures" on the background clustering of graphs. "Normal pictures" are pictures of normal busines...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06F16/55
CPCG06F16/55
Inventor 田春霖蒋泽锟严宋扬阮书宁
Owner 安徽深信科创信息技术有限公司