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

Fish identification method based on multi-perspective acoustic data

An identification method and multi-view technology, which are applied in the re-radiation of sound waves, the use of re-radiation, measurement devices, etc., to achieve the effect of simple structure and convenient realization.

Active Publication Date: 2015-04-22
HARBIN ENG UNIV
View PDF5 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to solve the problem of high-precision fish identification based on acoustic echo data, and provide a fish identification method based on multi-view acoustic data

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
  • Fish identification method based on multi-perspective acoustic data
  • Fish identification method based on multi-perspective acoustic data
  • Fish identification method based on multi-perspective acoustic data

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0011] Below in conjunction with accompanying drawing and implementation the present invention will be further introduced:

[0012] figure 2 It is a schematic diagram of multi-view data acquisition method, wherein the middle black point represents the transmitting transducer, and the gray points on both sides represent the receiving transducer. The transmitting transducer of the present invention is a narrowband phased array transducer, and the receiving transducer is a standard water transducer. hearing device.

[0013] In the present invention, the combination of wavelet packet decomposition and discrete cosine transform is used for feature extraction, and the feature extraction process is as follows image 3 shown. Feature extraction mainly includes several steps such as wavelet packet transform, discrete cosine transform, and feature dimensionality reduction. The sub-band energy contains the response information of fish to different frequency bands, and the discrete cos...

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 fish identification method based on multi-perspective acoustic data. A single-beam phase-control energy transducer is used to emit acoustic signals to a fish body, M hydrophones receive scattered acoustic signals of the fish body in M directions, the obtained multi-perspective scattered signals are filtered and normalized, a wavelet packet is decomposed and reconstructed, sub-band energy of the wavelet packet is solved, feature extraction is carried out in the discrete cosine transform manner to obtain the transformation ratio, the characteristic quantity composed of the sub-band energy of the wavelet packet and the transformation ratio are input to a first-stage classifier to make decision, a posterior probability vector output in each direction is obtained from the decision result in a probability estimation formula, the posterior probability vectors in the directions are combined to obtain a multi-perspective decision posterior probability, the multi-perspective decision posterior probability is input to a second-stage support vector machine, and a final classification result is output to obtain the fish identification method. The problems that in the prior art, scattered information of the fish can be only detected in a single direction and data loss is caused are solved.

Description

technical field [0001] The present invention relates to an acoustic fish recognition method, more specifically to a fish recognition method based on multi-view acoustic data. Background technique [0002] 201210063846.0 A method and system for fish identification based on multi-scale information entropy of wavelet packets. A fish identification technology based on acoustic methods is introduced. Fish is a complex sound scatterer. The orientation, size and shape of fish body and swim bladder It seriously affects the acoustic scattering characteristics of fish, so that the acoustic scattering signal of fish becomes very complex. Extracting features that can reflect the intrinsic acoustic scattering characteristics of fish from complex signals is particularly important in fish identification. However, for fish species with swim bladders, the swim bladder is the main contributor to the scattering signal. The swim bladders of different fish species are different in size and shape...

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): G01S15/96G01S7/539
CPCG01S7/539G01S15/88
Inventor 杜伟东李海森魏玉阔徐超朱建军周天陈宝伟
Owner HARBIN ENG UNIV