Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Sample extraction and expansion method for small sample image recognition, and storage medium

A technology of image recognition and sample expansion, applied in the field of image processing, can solve the problems of generating wrong samples and affecting the training effect, and achieve the effect of stable effect, improved recognition effect and accurate data

Active Publication Date: 2021-08-24
HARBIN ENG UNIV
View PDF10 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to solve the problem of generating wrong samples that may be caused by the method of generating new samples in the process of small-sample image recognition, thereby affecting the training effect

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
  • Sample extraction and expansion method for small sample image recognition, and storage medium
  • Sample extraction and expansion method for small sample image recognition, and storage medium
  • Sample extraction and expansion method for small sample image recognition, and storage medium

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0050] This embodiment is a sample extraction method for small-sample image recognition, which is essentially a sampling method based on feature reconstruction. Its main idea is to optimally divide large-sample data through unsupervised fuzzy clustering. Then reconstruct the typical small sample data set. Specifically, the following steps are included:

[0051] First, calculate the central support point C of each category of the image large sample data set k .

[0052] Secondly, from the perspective of sample characteristics, the same kind of sample data is divided into dynamic number of clusters, and the centroid of each cluster is calculated according to each division situation.

[0053] Then, according to the centroid of each cluster, calculate the new center point of the category in this division. And by calculating the sum of the error of the same cluster and the error of the centroid as the total error of the category in this division. The case with the smallest erro...

specific Embodiment approach 2

[0076] This embodiment is a storage medium, and at least one instruction is stored in the storage medium, and the at least one instruction is loaded and executed by a processor to implement a sample extraction method for small-sample image recognition.

specific Embodiment approach 3

[0078] This embodiment is a sample expansion method for small-sample image recognition, which is essentially a sample expansion method based on intra-class deformation information. Its main idea is to generate and expand a new data set of this category by learning the deformation information between the same category of data, and then using the deformation information on other samples of the same category. In order to better learn the deformation information between similar data, through the sample extraction method of feature reconstruction, the specific implementation mode 1 provides a typical small-sample data set with relatively complete features for data expansion, and each category in the typical small-sample data set All are divided into multi-clusters with characteristic differences, and then expanded based on the small sample data set determined in the first embodiment. figure 1 It is an overview diagram of the sample extraction and expansion method based on feature r...

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 sample extraction and expansion method for small sample image recognition, and a storage medium, and belongs to the technical field of image processing. The objective of the invention is to solve the problem of generation of wrong samples possibly caused by adoption of a new sample generation mode in a small sample image recognition process. The invention firstly provides a sample extraction method based on feature reconstruction to solve the problem of feature missing of a small sample data set, and realizes extraction of a typical small sample data set from a large sample data set from the perspective of data features. According to the method, the centroid of the large sample data is used as a standard of extraction measurement, so that the extracted typical small sample data set has more comprehensive characteristics, and the effect is more stable. The invention further provides a sample expansion method based on the deformation information, and the extracted typical small sample data set is expanded into a new large sample data set by utilizing the deformation information between data of the same type and different clusters in optimal division. The method is mainly used for sample extraction and expansion of small sample image recognition.

Description

technical field [0001] The invention relates to an image sample extraction and expansion method and a storage medium. It belongs to the technical field of image processing. Background technique [0002] Neural networks often require a large amount of data to complete effective training. In the low data area, the training effect and generalization ability of the network will be poor. [0003] In order to ensure the recognition effect of the network, the small-sample learning method based on data expansion is basically used at present. Most of the data expansion methods use the idea of ​​generative confrontation to generate data or directly use the difference information between similar data sets to achieve data enhancement. However, this method does not consider whether the features of the small sample data set are complete before expansion, which will cause the data after expansion to still lack important features. There is also no question of whether the use of differenc...

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): G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F18/241G06F18/214
Inventor 王红滨张政超张耘王念滨周连科张毅湛浩旻
Owner HARBIN ENG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products