Complex picture clustering method based on adaptive weight

A technique of self-adaptive weight and clustering method, which is applied in the fields of still image data clustering/classification, neural learning method, still image data retrieval, etc.

Pending Publication Date: 2021-09-28
INST OF ELECTRONICS & INFORMATION ENG OF UESTC IN GUANGDONG
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Based on the above problems, the present invention provides a complex image clustering method based on adaptive weights, which solves the problem that in the existing image clustering model, the quality of graphic samples cannot be used to determine the weight of a sample, and the adaptive weight Problems with model training

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
  • Complex picture clustering method based on adaptive weight
  • Complex picture clustering method based on adaptive weight
  • Complex picture clustering method based on adaptive weight

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0028] The present invention will be further described below in conjunction with the accompanying drawings. Embodiments of the present invention include, but are not limited to, the following examples.

[0029] Such as figure 1 A complex image clustering method based on adaptive weight is shown, including the following steps:

[0030] Step 1: Construct an image data set, divide the image data set into a training set and a verification set, and preprocess the image.

[0031] In this step, according to 90% training set and 10% verification set, the image data set is divided into training set and verification set. Each class contains 100 samples, and the sample size is not uniform.

[0032] In this step, the image preprocessing includes upsampling and downsampling operations on the image, wherein the upsampling adopts the cubic interpolation method, and the downsampling is based on the target size and the original size. The image is subjected to upsampling and downsampling Af...

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 relates to the field of deep learning and image clustering, in particular to a complex image clustering method based on adaptive weight, which comprises the following steps of: firstly, initializing a clustering network by using an existing classification network and a traditional clustering algorithm; secondly, clustering the images, calculating a network target of the iteration in the entropy reduction direction, and updating the network; thirdly, calculating the weight of each sample in the next iteration by using the sample entropy; and finally, when the clustering loss is smaller than an iteration stopping threshold value, outputting a clustering result. The problem that in an existing image clustering model, it is difficult to determine the weight of one sample by using the quality of the graphic sample and carry out model training by using the self-adaptive weight is solved.

Description

technical field [0001] The invention relates to the field of deep learning and image clustering, in particular to a complex image clustering method based on adaptive weights. Background technique [0002] With the popularization of portable media devices, images are generated faster and faster. Because of their intuitive and rich content display methods, images have become one of the most important resources in the era of big data; for image-based work, including object recognition, product recommendation and other applications , image clustering is its underlying work; the clustering work of traditional simpler images such as license plate numbers has been well developed, and for more and more complex images, existing image clustering algorithms Even deep image clustering algorithms cannot adapt to it well; complex image classification networks have been well developed, such as ResNet, GoogleNet, etc., but it is impossible to accurately label massive images, so for complex ...

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): G06F16/55G06K9/62G06N3/08
CPCG06F16/55G06N3/084G06F18/2321G06F18/214
Inventor 任亚洲杨之蒙吴子锐
Owner INST OF ELECTRONICS & INFORMATION ENG OF UESTC IN GUANGDONG
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap