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

A One-dimensional Image Recognition Method Based on Particle Swarm Optimization and Deep Learning Feature Selection

A particle swarm optimization and deep learning technology, applied in neural learning methods, scene recognition, character and pattern recognition, etc., can solve problems such as high feature dimension, gradient dispersion, and increase in running time, so as to solve the problem of excessive feature dimension. High, reduce impact, improve the effect of recognition rate

Active Publication Date: 2022-04-22
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF9 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, these methods also have some problems. The feature dimension extracted by the deep neural network is usually relatively high, which will reduce the performance of subsequent classifiers.
Secondly, when the depth of the network is deepened, the problem of gradient dispersion will occur, and the running time will also be greatly increased.

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
  • A One-dimensional Image Recognition Method Based on Particle Swarm Optimization and Deep Learning Feature Selection
  • A One-dimensional Image Recognition Method Based on Particle Swarm Optimization and Deep Learning Feature Selection
  • A One-dimensional Image Recognition Method Based on Particle Swarm Optimization and Deep Learning Feature Selection

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0055]In order to verify the effectiveness of the proposed method, the following simulation experiments are carried out.

[0056] The training sample data and test sample data used in the experiment are all from the data of five aircraft (ah-64, an-26, b-1b, b-52, f-15) simulated by a radar aircraft target simulation software. The pitch angle of the aircraft relative to the radial direction of the radar is 0°, and the azimuth angle range: 0°-180°. Each set of aircraft data has 1800 one-dimensional images, and each one-dimensional range image has 320 range units. The working parameters of the radar are: signal bandwidth: 400MHz, center frequency: 6GHz, 1200 aircraft data are randomly selected from each group as the training set, and the remaining 600 are used as the test set.

[0057] In this experiment, two kinds of networks, deep belief network and stacked autoencoder, were compared with the feature selection method of particle swarm deep learning, and BP network was used as...

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 belongs to the technical field of radar target recognition, and in particular relates to a one-dimensional image recognition method based on particle swarm optimization deep learning feature selection. The method of the present invention first uses the deep belief network (DBN) to learn the internal characteristics of the original high-dimensional data, and further combines the particle swarm optimization (PSO) algorithm to realize the selection and optimization of the characteristics and solve the problem of excessively high dimensionality of the characteristics of the deep learning network. Problems, reduce the impact of redundant features on subsequent classifiers, and improve the recognition rate of the target.

Description

technical field [0001] The invention belongs to the technical field of radar target recognition, and in particular relates to a one-dimensional image recognition method based on particle swarm optimization deep learning feature selection. Background technique [0002] Radar target recognition technology based on radar one-dimensional range profile is an important radar target recognition technology in recent years. The radar one-dimensional range image is the vector sum of the projected echoes of target scattering points acquired by broadband radar along the radar line of sight, which contains a wealth of information such as the geometric structure of the target range image. Compared with SAR and ISAR images, a The three-dimensional range image is easier to obtain and process, so it is widely used in the field of radar target recognition. [0003] Deep learning is a recognition method proposed in recent years. It has made good progress in many fields, such as image recognit...

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 Patents(China)
IPC IPC(8): G06V20/13G06V10/771G06V10/764G06V10/82G06N3/08G06N3/00
CPCG06N3/084G06N3/006G06V20/13
Inventor 周代英黎晓烨赖陈潇李雄冯健
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA
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