Structured sparse feature extraction for underwater targets

A sparse feature, underwater target technology, applied in radio wave measurement systems, instruments, etc., can solve problems such as noise interference

Active Publication Date: 2017-06-13
NORTHWESTERN POLYTECHNICAL UNIV
View PDF7 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

These target features have good recognition performance under certain conditions, but they are also limited by various practical application conditions such as noise interference.

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
  • Structured sparse feature extraction for underwater targets
  • Structured sparse feature extraction for underwater targets
  • Structured sparse feature extraction for underwater targets

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment 1

[0058] Step 1: Select a data set containing 3 types of targets, a total of 45 sample files in wav format, and 15 for each type. The duration of each sample file varies from 5 to 6 seconds, and the sampling frequency is 8000Hz. Read the wav data samples, use MATLAB software to preprocess the data samples, and perform frame processing. The duration of each frame is 0.25s, that is, 2000 data points, and there is a partial overlap of 1 / 3 frame length between adjacent frames. Finally, the direct current component is removed for each frame sample, and the energy is normalized to [0,1] to eliminate the influence of absolute size on the classification recognition effect.

[0059] Step 2: Select an appropriate frequency band and frequency resolution, and construct a discrete Fourier dictionary. The number of rows of the dictionary is consistent with the length of each frame sample, and the number of columns is determined according to the frequency range and number of line spectrum co...

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 method for extracting the characteristics of underwater targets radiated noise based on Bayesian structured sparse. The method for extracting the characteristics of underwater target radiated noise comprises the following steps: firstly, the radiated noise signals of underwater targets are divided into frames, and the hierarchical Bayesian model is used to model the frame signals based on the discrete Fu Liye dictionary; for the adjacent multi frame signals, the Bayesian variational algorithm is used to deduce the model, and the decomposition coefficients of the signals are estimated; finally, the normalized decomposition coefficients are taken as frame structured sparse features of frame signals. The characteristics of underwater targets radiated noise is a kind of target characteristic that is robust to noise.

Description

technical field [0001] The invention belongs to the field of underwater target identification, and is used for extracting features from noise signals radiated by targets and applied to target classification or identification. Background technique [0002] Underwater target recognition is an important function of modern sonar systems and underwater acoustic countermeasures systems, and it is mainly done manually by sonar personnel at present. However, the training of sonarists requires a lot of time and capital costs, and the actual performance of sonarists is easily affected by physiological, psychological and environmental factors. With the automation and intelligence of modern sonar systems and underwater acoustic countermeasure systems, automatic underwater target recognition technology that does not rely on artificial intelligence has become an important research content in the field of underwater target recognition, which has great practical and long-term strategic sign...

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): G01S7/539
CPCG01S7/539
Inventor 王璐曾向阳其他发明人请求不公开姓名
Owner NORTHWESTERN POLYTECHNICAL UNIV
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