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

Deep convolutional network-based ship noise identification and classification method

A recognition classification, deep convolution technology, applied in neural learning methods, character and pattern recognition, biological neural network models, etc., can solve the problems of confusion in recognition, poor recognition effect, and inconspicuous difference characteristics, and achieve the authenticity of the effect. , has the effect of universality and reduced intervention

Inactive Publication Date: 2018-01-19
HARBIN ENG UNIV
View PDF0 Cites 20 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Therefore, it becomes very difficult to use the neural network to process files, and the difference features extracted by the three-layer network will not be obvious, which will lead to recognition confusion and make the recognition effect very poor

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
  • Deep convolutional network-based ship noise identification and classification method
  • Deep convolutional network-based ship noise identification and classification method
  • Deep convolutional network-based ship noise identification and classification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031] Below in conjunction with accompanying drawing, the method proposed by the present invention is described in further detail:

[0032] At present, experts in the field of underwater target noise are actively researching, and scholars have extracted many models and methods. It is hoped that a universally applicable model can be found, and the processing process does not require too much human intervention. However, for the existing model, it is difficult to achieve these two points. First, for BP, this model is not suitable for processing sound files with a large amount of data, and the network layer is relatively shallow, and the feature extraction is not complete enough. For other models, human intervention is required. The influence of subjective factors is relatively large. This model analyzes the traditional model, improves it according to the shortcomings of the traditional model, and proposes a model algorithm based on MFCC and depth convolution.

[0033] This i...

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 a deep convolutional network-based ship noise identification and classification method, and belongs to the field of underwater ship noise identification. A solution is proposedmainly for the problems of small file processing quantity, unremarkable extracted features and easy falling into local optimal solution in a BP neural network. According to the method, firstly according to MFCC, noises are removed in an original sound; corresponding effective features are extracted; and in the process, the noises with high interference are removed. A sound file processed throughthe MFCC is converted to be in a format receivable for a deep convolutional network. The effective features are extracted in a multilayer manner through the deep convolutional neural network; and theextracted features are higher in validity and higher in universality. A large amount of sound data in reality is identified and classified, so that the manual intervention degree is reduced, differentship noises can be better distinguished, and the purpose of identification is achieved.

Description

technical field [0001] The invention relates to a ship noise recognition and classification method based on a deep convolutional network, which belongs to the field of underwater ship noise recognition. Background technique [0002] Ship noise is divided into air noise (noise generated in the air medium) and water noise (noise generated in the water medium). Among them, ship water noise is divided into ship radiation noise and ship self-noise. Ship radiation noise Noise is generated by mechanical operation and ship movement on ships and radiates into the water. Ship radiated noise is the information source of passive detection devices and one of the important indicators of ship concealment. Mechanical noise and propeller noise constitute the main radiated noise. The importance of these two kinds of noise depends on the speed and depth of the ship and the frequency of the noise. Hydrodynamic noise has a great influence on self-noise. Generally, once the propeller cavitates, ...

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): G06K9/00G06N3/04G06N3/08G06F17/14G10L21/0208
Inventor 王念滨朱洪瑞何鸣顾正浩李浩然赵新杰王昆明苏畅童鹏鹏仝彤
Owner HARBIN ENG UNIV
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