Improved semi-supervised image classification method, equipment and storage medium

A classification method and semi-supervised technology, applied in neural learning methods, instruments, biological neural network models, etc., can solve the problems of uncontrollable models, difficult convergence, model collapse, etc., to improve the discrimination ability, low error rate, and good testing. The effect of error rate

Pending Publication Date: 2021-11-26
XINJIANG INFORMATION IND
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, it also has many shortcomings, such as mode collapse, uncontrollable model, and difficulty in convergence during training, etc.

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
  • Improved semi-supervised image classification method, equipment and storage medium
  • Improved semi-supervised image classification method, equipment and storage medium
  • Improved semi-supervised image classification method, equipment and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0079] An improved semi-supervised image classification method based on Triple-Gan, comprising the following steps:

[0080] (1) Build an improved semi-supervised image classification model; as figure 1 As shown, the improved semi-supervised image classification model includes a generator, a discriminator and a set of sub-judgers, and a set of sub-judgers includes the first sub-judger D s1 and the second sub-determinator D s2 ;

[0081] The generator G draws from the distribution P z Extract random noise z as input, and train simultaneously with all judgers in a way of generating confrontation; judger D 0 is the discriminative part of the improved semi-supervised image classification model, which divides the input image into the correct category as much as possible; a set of sub-judgers includes the first sub-judger D s1 and the second sub-determinator D s2 , except for the last few layers, they and the discriminator D 0 There are roughly the same network layers and para...

Embodiment 2

[0149] A computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of the Triple-Gan-based improved semi-supervised image classification method described in Embodiment 1 when executing the computer program.

Embodiment 3

[0151] A computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the Triple-Gan-based improved semi-supervised image classification method described in Embodiment 1 are implemented.

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 an improved semi-supervised image classification method, equipment and a storage medium. The method comprises the following steps: (1) constructing an improved semi-supervised image classification model, wherein the improved semi-supervised image classification model comprises a generator, a judger and a plurality of sub-judgers (secondary judgers); (2) training the improved semi-supervised image classification model; and (3) inputting an image to be classified into the improved semi-supervised image classification model trained in the step (2) to obtain a classification result. The improved semi-supervised image classification model shows a good test error rate, and the classification capability can be better improved by using unmarked images and generated images.

Description

technical field [0001] The invention belongs to the technical field of deep learning artificial intelligence, and in particular relates to an improved semi-supervised image classification method, equipment and storage medium based on Triple-Gan. Background technique [0002] In recent years, artificial intelligence has been rapidly developed and applied in various fields. Most methods in machine learning and deep learning belong to statistics, and statistics are also the theoretical support of artificial intelligence. Starting from solving statistical problems, a large number of The data is collected and processed, and then the possible results of the target data are arranged, combined and estimated to realize the simulation of the real results. According to the number of labels in the collected data, the classification task can be divided into three directions, namely the classification of supervised learning, the classification of semi-supervised learning and the classific...

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): G06K9/62G06N3/08
CPCG06N3/08G06F18/22G06F18/2415G06F18/214
Inventor 邹帅钱涛余长江邱斌钟方伟苟文清周莉
Owner XINJIANG INFORMATION IND
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