Check patentability & draft patents in minutes with Patsnap Eureka AI!

Intelligent grouping method and grouping system

A grouping method and grouping system technology, applied in the field of intelligent grouping method and grouping system, can solve the problems of high algorithm time complexity and unfavorable application, and achieve the effects of increasing feature matching accuracy, reducing quantization error, and refining data grouping results.

Pending Publication Date: 2022-06-24
广州炫视智能科技有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, it has been found in practice that the existing grouping methods all use the nearest neighbor feature matching between two images, or use global feature clustering to group images. When the image size is large, the time of these algorithms is complicated. are very high, which is not conducive to practical application

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
  • Intelligent grouping method and grouping system
  • Intelligent grouping method and grouping system
  • Intelligent grouping method and grouping system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0060] see figure 1 , figure 1 It is a schematic flowchart of an intelligent grouping method disclosed in an embodiment of the present invention. like figure 1 As shown, the intelligent grouping method may include the following steps.

[0061] 101. The grouping system calculates the distance from the feature vector of each data node in the data node set to the cluster center of each group of classes.

[0062] As an optional implementation, in this embodiment of the present invention, the cluster center is the center point position in each group of classes, and the present application can vectorize data nodes and cluster centers in advance to greatly improve the data Efficiency of grouping, for example, assuming that points A to D represent cluster centers, and points 1, 2, and 3 represent feature vectors of data nodes, the application can calculate the distance from point 1 to point A to D, and point 2 to point The distance from A to D, the distance from point 3 to point A...

Embodiment 2

[0072] see figure 2 , figure 2 It is a schematic flowchart of another intelligent grouping method disclosed in an embodiment of the present invention. like figure 2 As shown, the intelligent grouping method may include the following steps:

[0073] 201. The grouping system calculates the distance from the feature vector of each data node in the data node set to the cluster center of each group of classes.

[0074] 202. The grouping system selects a first target data node whose distance from the same cluster center is less than a specified threshold.

[0075] 203. The grouping system assigns the first target data node to a group class corresponding to the same cluster center.

[0076] 204. The grouping system detects whether there are any unallocated remaining data nodes in the data node set, and if so, executes steps 205 to 209, and if not, ends this process.

[0077] As an optional implementation manner, in this embodiment of the present invention, the present applica...

Embodiment 3

[0101] see image 3 , image 3 It is a schematic structural diagram of a grouping system disclosed in an embodiment of the present invention. like image 3 As shown, the grouping system 300 may include a first calculating unit 301, a selecting unit 302 and an allocating unit 303, wherein:

[0102] The first calculation unit 301 is configured to calculate the distance from the feature vector of each data node in the data node set to the cluster center of each group of classes.

[0103] The selection unit 302 is configured to select the first target data node whose distance from the same cluster center is smaller than a specified threshold.

[0104] The assigning unit 303 is configured to assign the first target data node to the group class corresponding to the same cluster center.

[0105] As an optional implementation, in this embodiment of the present invention, the cluster center is the center point position in each group of classes, and the present application can vecto...

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 embodiment of the invention relates to the technical field of artificial intelligence, and discloses an intelligent grouping method and system, and the method comprises the steps: calculating the distance from the feature vector of each data node in a data node set to the clustering center of each group class; selecting a first target data node, wherein the distance between the first target data node and the same clustering center is smaller than a specified threshold value; and distributing the first target data node to a group class corresponding to the same clustering center. By implementing the embodiment of the invention, the quantization error can be effectively reduced, and the feature matching precision can be increased, so that a more refined data grouping result can be obtained.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to an intelligent grouping method and a grouping system. Background technique [0002] With the rapid development of optoelectronic technology and the Internet, people can easily obtain image data of large-scale scenes, so the demand and application of recovering 3D scene structure from large-scale image collections have emerged as the times require. However, due to the different sensor parameters used to acquire images, as well as different shooting fields and angles, the image collection not only contains multiple images with different scene contents, but also contains noise pollution, blurred jitter, and even erroneous images. Therefore, preprocessing the disordered image collection and grouping the images related to the content not only helps the user to quickly organize and grasp the image content, but also is the precondition and key step for the 3D reconstruct...

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 Applications(China)
IPC IPC(8): G06V10/764G06V10/762G06V10/74G06K9/62
CPCG06F18/23G06F18/22G06F18/24
Inventor 吴振涛刘辉赵鲜
Owner 广州炫视智能科技有限公司
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com