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

Mineral image data expansion method and device based on generative adversarial network

An image data and expansion device technology, applied in biological neural network models, neural learning methods, instruments, etc., can solve problems such as difficult to meet, and achieve the effects of easy operation, convenient deployment, and balanced data distribution

Pending Publication Date: 2022-05-31
KUNMING UNIV OF SCI & TECH
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

With the development of computer technology, the model determination method using machine learning technology will gradually replace naked eye identification and become the mainstream of future development. Most of the methods based on deep learning have higher requirements for the amount of labeled data, which is often difficult to meet in real scenarios.

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
  • Mineral image data expansion method and device based on generative adversarial network
  • Mineral image data expansion method and device based on generative adversarial network
  • Mineral image data expansion method and device based on generative adversarial network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0031] Example 1: as Figure 1-3 As shown, a method for expanding mineral image data based on a generative adversarial network includes: obtaining a text vector according to a text description; and obtaining a mineral image according to the text vector.

[0032] Further, the text vector can be obtained by setting the description according to the text, including:

[0033] Obtain the existing mineral image, and describe the existing mineral image in text according to the existing mineral image features;

[0034] Perform feature modification on the existing text description, and use the modified text description as the text information to be converted;

[0035] Input the text information to be converted into the RNN network to obtain text vectors containing mineral feature information.

[0036] Further, the feature may be set to represent a direct observation feature, and the direct observation feature includes at least one of the following: color feature, texture feature, morp...

Embodiment 2

[0064] Embodiment 2: According to another aspect of the embodiments of the present invention, a device for expanding mineral image data based on a generative adversarial network is provided, including:

[0065] Text information vectorization module, used to obtain text vector according to text description;

[0066] The text generation image module is used to obtain mineral images according to the text vector.

[0067] Further, the text information vectorization module can be set, including:

[0068] a first preprocessing unit, used for acquiring an existing mineral image, and textually describing the existing mineral image according to the characteristics of the existing mineral image;

[0069] The second preprocessing unit is used to perform feature modification on the existing text description, and use the modified text description as the text information to be converted;

[0070] The obtaining unit is used to input the text information to be converted into the RNN network...

Embodiment 3

[0071] Embodiment 3: According to another aspect of the embodiments of the present invention, a processor is provided, and the processor is configured to run a program, wherein when the program runs, execute any one of the generative adversarial network-based methods described above. Mineral Image Data Augmentation Methods.

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 mineral image data expansion method and device based on a generative adversarial network, the mineral image data expansion method and device based on the generative adversarial network are convenient to deploy and easy to operate, specifically, an image is generated on the basis of an original image, the possible symbiosis situation of some minerals can be simulated, and then the CGAN network is further used, so that the mineral image data expansion efficiency is improved. Compared with the characteristics that original data distribution is not changed, the definition of the image is also influenced to a certain degree and a balancing effect on a data set with unbalanced sample distribution cannot be achieved by means of cutting, overturning, rotating and the like, the means adopted by the method is not limited by rules, the distribution of the original data can be changed to a certain degree, and the method has the advantages of being high in practicability and the like. And meanwhile, the definition of the image is not influenced, and data with a small number of samples can be expanded, so that the overall data distribution is balanced. Based on the mineral image obtained by the method, the problem that a network model is difficult to train due to too little mineral image data can be solved.

Description

technical field [0001] The invention relates to a mineral image data expansion method based on a generative confrontation network, and relates to the data processing field of geological exploration and mineral processing. Background technique [0002] The mineral industry plays an important role in economic construction. So far, minerals have been widely used in various fields of industrial and agricultural production and people's life. [0003] In the process of mineral exploration, the rapid and accurate identification of rock-forming minerals is of great significance to determine the quality and reserves of mineral resources and to determine the scale of mining. Mineral identification can be divided into naked eye identification and instrument identification according to whether the instrument is used, and naked eye identification is the basis of instrument identification. Because through the naked eye identification, the species or family of minerals can be preliminaril...

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): G06K9/62G06N3/04G06N3/08G06V10/80G06V10/82
CPCG06N3/08G06N3/045G06F18/253
Inventor 杨彪马亦骥成宬王彬胡蓉朱荣钱斌李琨
Owner KUNMING UNIV OF SCI & TECH
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

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com