Unlock instant, AI-driven research and patent intelligence for your innovation.

Acoustic Haar feature extraction method and system for underwater target recognition

A technology for underwater target and feature extraction, applied in character and pattern recognition, instruments, computer parts, etc., can solve the problem of low recognition accuracy, achieve strong practicability and improve training efficiency

Inactive Publication Date: 2022-02-01
INST OF ACOUSTICS CHINESE ACAD OF SCI
View PDF12 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

After feature extraction, a classifier must be selected for recognition. Due to the nonlinearity of the target signal in water, the traditional linear-based recognition method has a relatively low recognition accuracy.

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
  • Acoustic Haar feature extraction method and system for underwater target recognition
  • Acoustic Haar feature extraction method and system for underwater target recognition
  • Acoustic Haar feature extraction method and system for underwater target recognition

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0056] like figure 1As shown, embodiments of the present invention proposes an acoustic HAAR feature extraction method for target recognition in water. The method first converts the first-dimensional raw target signal after the pre-treated generates an acoustic time-frequency feature and extracts acoustic characteristics based on Haar-LIKE. Since the extraction is huge, it is difficult to use the training of classification identification algorithm. It is proposed that the feature screening algorithm based on Adaboost is proposed, and the feature combination for identifying representative, and the feature combination after screening constitutes an acoustic haar feature vector, this feature The input of the subsequent classifier is used for training and target recognition of the classifier.

[0057] In the first step, the target signal in the original one-dimensional water is changed to the two-dimensional acoustic time-frequency feature.

[0058] The acoustic time flow diagram cont...

Embodiment 2

[0086] Embodiment 2 of the present invention proposes an acoustic HAAR feature extraction system for the target recognition in water, based on the method of Embodiment 1, the system includes: a pre-processing module, a time-frequency feature conversion module, a mixing acoustic HAAR feature extraction module And significant acoustic feature extraction modules;

[0087] The pretreatment module is used to prepare the target water acoustic signal in the received water;

[0088] The time-frequency feature conversion module is configured to time-frequency feature conversion of the pre-treated signal, generate an acoustic video characteristic;

[0089] The hybrid acoustic HAAR feature extraction module is used to characterize the acoustic time-frequency characteristics based on Haar-Like features.

[0090] The significant acoustic feature extraction module is used to complete a significant acoustic feature extraction based on the AdapBoost algorithm.

[0091] The present invention uses ...

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 the field of underwater target feature extraction, in particular to an acoustic Haar feature extraction method and system for underwater target recognition, and the method comprises the steps: carrying out the preprocessing of a received underwater acoustic signal of an underwater target; performing time-frequency feature transformation on the preprocessed signal to generate an acoustic video feature map; on the basis of Haar-like features, carrying out feature extraction on the acoustic time-frequency feature map to obtain mixed acoustic Haar features; and completing saliency acoustic feature extraction based on an AdapBoost algorithm. The method is oriented to engineering application, an acoustic Haar feature extraction method is provided for recognition of the underwater target, the extracted saliency features are beneficial to improving the training efficiency, and the method has the advantages of being simple, easy to use, clear and high in practicability.

Description

Technical field [0001] The present invention relates to the field of underwater target feature extraction, and more particularly to an acoustic HAAR feature extraction method and system for target recognition in water. Background technique [0002] The water sound target recognition technology is a passive target radiation noise, active target echo, and other sensor information extracting target features and discriminating target types or category information processing techniques. Whether in civil or military fields, water sound target identification has an extremely important application background. However, due to the complexity of the marine environment and the particularity of the water sound signal channel, it is necessary to reflect the characteristics of the target nature, but also meet the requirements of the target radiation noise signal, and can meet the requirements of the target detection. difficulty. [0003] The acoustic feature extraction of the target signal of t...

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): G06V10/52G06V20/05G06V10/764G06K9/62
CPCG06F18/24
Inventor 孙飞虎冯靖钊高善国李宇
Owner INST OF ACOUSTICS CHINESE ACAD OF SCI