Underwater target recognition method based on improved PSO-TSNE feature selection

A PSO-TSNE, underwater target technology, applied in the field of image recognition, can solve the problems of reduced recognition rate and high data dimension

Pending Publication Date: 2021-04-09
FUJIAN UNIV OF TECH
View PDF5 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In order to ensure the correct rate of recognition, multiple features should be combined, but ...

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
  • Underwater target recognition method based on improved PSO-TSNE feature selection
  • Underwater target recognition method based on improved PSO-TSNE feature selection
  • Underwater target recognition method based on improved PSO-TSNE feature selection

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0048] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0049] Such as Figure 1 to Figure 3 As shown in one of them, the present invention discloses an underwater target recognition method based on improved PSO-TSNE feature selection, which comprises the following steps:

[0050] Step 1, the underwater target radiation noise signal is subjected to underwater noise feature extraction and the data is normalized to obtain a standardized data set;

[0051] Step 2, use T-SNE to perform dimensionality reduction processing on the data of the standardized data set, and obtain the high-dimensional spatial data point x i with x j The similarity between p ij , data point y in low-dimensional space iwith y j The simil...

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 an underwater target recognition method based on improved PSO-TSNE feature selection. The underwater target recognition method comprises the steps of performing underwater noise feature extraction on an underwater target radiation noise signal and normalizing data to obtain a standardized data set; performing dimension reduction processing on the data of the standardized data set by using T-SNE to obtain similarity between high-dimensional space data points, similarity between data points in a low-dimensional space, and an objective function and a step function for KL divergence optimization of the high-dimensional space data points and the low-dimensional space data points; obtaining an optimal KL value by using an improved PSO algorithm; and carrying out dimension reduction processing on the data of the standardized data set by using the minimum KL and the T-SNE again, and taking underwater target features to confirm a target. According to the method, the updated PSO is adopted to optimize the divergence in the T-SNE, so that the performance of the T-SNE is better.

Description

technical field [0001] The invention relates to the technical field of image recognition, in particular to an underwater target recognition method based on improved PSO-TSNE feature selection. Background technique [0002] Underwater target recognition is one of the technical problems to be solved urgently in the field of underwater acoustic technology. Image processing and recognition of underwater targets is an indispensable technical guarantee for underwater robots to work normally. The feature extraction of the target noise signal is the key technology of the target recognizer, so scholars have proposed a variety of feature extraction methods, trying to obtain the characteristics of the noise signal from different angles. In order to ensure the correct rate of recognition, multiple features should be combined, but this will bring about the problem that the data dimension is too high and the recognition rate will decrease. Contents of the invention [0003] The purpos...

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
IPC IPC(8): G06K9/62G06K9/46G06N3/00
CPCG06N3/006G06V10/44G06F18/2135G06F18/22
Inventor 孟振宇钟于心
Owner FUJIAN UNIV OF TECH
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