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

Sunspot identification method based on deep learning

A sunspot and deep learning technology, applied in the field of sunspot identification, can solve problems such as missed identification and false identification

Inactive Publication Date: 2018-04-06
KUNMING UNIV OF SCI & TECH
View PDF7 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The present invention provides a method for identifying sunspots based on deep learning. The method of fully convolutional neural network in deep learning is applied to identifying sunspots, which is used to solve the misidentification and missing identification of traditional methods for identifying sunspots. The problem

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
  • Sunspot identification method based on deep learning
  • Sunspot identification method based on deep learning
  • Sunspot identification method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0023] Embodiment 1: as Figure 1-5 As shown, a method for identifying sunspots based on deep learning first uses a morphological erosion and expansion method to extract sunspot data samples, and then makes label information and LMDB data sets through the extracted sunspot samples; then designs convolution The content and parameters of the configuration file of the neural network, train and test the convolutional neural network model including the data layer, convolutional layer and fully connected layer, convert the fully connected layer of the network into a convolutional layer, and then scale the full sun image Then input it into the converted full convolutional neural network to calculate the probability, and finally screen out the sunspots that meet the threshold, and use non-maximum value suppression to mark the sunspot position on the full sun surface image.

[0024] The specific steps of the method for identifying sunspots based on deep learning are as follows:

[002...

Embodiment 2

[0033] Embodiment 2: as Figure 1-5 As shown, a method for identifying sunspots based on deep learning, first preprocesses the full sun image, then extracts sunspot data samples, and makes label information and LMDB data sets through the extracted sunspot samples; then designs convolution The configuration file of the neural network, train and test the convolutional neural network model including the data layer, the convolutional layer and the fully connected layer, convert the fully connected layer of the network into a convolutional layer, and then scale the full sun image and input it to the conversion The probability is calculated in the final full convolutional neural network, and finally the sunspots that meet the threshold are screened out, and the sunspot position is marked on the full sun surface image by non-maximum suppression.

[0034] The specific steps of the method for identifying sunspots based on deep learning are as follows:

[0035] Step1, image preprocessi...

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 a sunspot identification method based on deep learning, and belongs to the field of astronomical technology, image processing and artificial intelligence. The sunspot identification method comprises the steps that firstly a sunspot data sample is extracted and tag information and an LMDB data set are made; then the configuration file of a convolutional neural network is set, a convolutional neural network model including a data layer, a convolutional layer and a full connection layer is trained and tested and the full connection layer of the network is converted into the convolutional layer; then the full sun-face image is proportionally scaled and then inputted to the converted full convolutional neural network to calculate the probability; and finally the sunspots meeting the threshold are selected out through screening, and the positions of the sunspots are marked on the full sun-face image by using non-maximum suppression. Application of the method of deeplearning to the problem of sunspot identification is unprecedented and effective, and the problems of misidentification and leak identification of the conventional sunspot identification method can besolved to a great extent.

Description

technical field [0001] The invention relates to a method for identifying sunspots based on deep learning, which belongs to the fields of astronomical technology, image processing and artificial intelligence. Background technique [0002] When the magnetic field on the surface of the sun changes drastically, it usually causes the earth's magnetic field to be disordered, which affects the life of living things on earth. The eruption of solar flares is a manifestation of violent changes in the magnetic field. According to related research, there is a close relationship between sunspot groups and flare eruptions. The identification of sunspots can provide a feasible technical means for the prediction of flares. Therefore, the sunspots on the full sun surface Correct identification is very important. [0003] At present, manual and automatic methods are mainly used in identifying sunspots. In the early stage, the manual method can only analyze a small number of images every day...

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): G06K9/46G06K9/62G06T7/136G06T7/77
CPCG06T7/136G06T7/77G06T2207/20081G06T2207/20084G06V10/44G06F18/214
Inventor 杨云飞付小娜廖成武杨洪娟张晓丽
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