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

Freshwater fish image real-time identification method based on lightweight convolutional network

A convolutional network and recognition method technology, applied in the field of real-time recognition of freshwater fish images, can solve the problems of difficult feature extraction, slow recognition, lack of data sets, etc., to achieve real-time recognition, a wide range, and reduce the amount of network parameters.

Pending Publication Date: 2021-06-15
XIDIAN UNIV
View PDF7 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to address the deficiencies in the prior art above, and propose a real-time recognition method for freshwater fish images based on lightweight convolutional networks, aiming to solve the problems of lack of data sets and difficult feature extraction in the existing freshwater fish recognition technology. , the problem of slow recognition speed

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
  • Freshwater fish image real-time identification method based on lightweight convolutional network
  • Freshwater fish image real-time identification method based on lightweight convolutional network
  • Freshwater fish image real-time identification method based on lightweight convolutional network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0024] The invention is further described below with reference to the drawings.

[0025] Refer figure 1 Further, the implementation steps of the present invention are further described.

[0026] Step 1. Build a freshwater fish image data set.

[0027] At least 2700 sizes of a freshwater fish image, all images cover at least 9 freshwater fish, and all of the present invention is currently a common freshwater fish, are squid, squid, grass carp, blue fish, Squid, squid, squid, squid, squid; freshwater fish image data mainly adopt high-definition camera shooting and reptile technology, shooting data accounts for main part, and the reptile technology acquires data as a secondary part.

[0028] The freshwater fish in each freshwater fish image is manually marked, and the species corresponding to the freshwater fish category, record the vertex coordinates of each of the external rectangular frames used in freshwater fish, and each picture corresponding to generate one Tag files in XML fo...

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 freshwater fish image real-time identification method based on a lightweight convolutional network, and aims to solve the problems of lack of data sets, difficulty in feature extraction and low identification speed in the existing freshwater fish species classification technology. The method comprises the following specific steps: (1) constructing a freshwater fish image data set; (2) constructing a lightweight deep convolutional neural network; (3) training a deep convolutional neural network; (4) identifying the freshwater fish test image in real time; by constructing the freshwater fish image data set and utilizing the trained lightweight deep convolutional neural network, real-time recognition is automatically performed on the collected freshwater fish image, and the method has the advantages of no need of manually extracting fish body features, high recognition precision, high speed and small hardware resource consumption.

Description

Technical field [0001] The present invention belongs to the field of image processing technology, and more, the real-time identification method of a freshwater fish image based on a lightweight convolutionary network in the field of image recognition. The present invention can be used in fishery monitoring, aquaculture, and leisure scenes for real-time monitoring and identification of captured freshwater fish. The identification results can be used for fish species collection, and can provide reference for rare fish species. Background technique [0002] During the traditional fish identification process, it is mainly used to identify the fish, and the automatic freshwater fish recognition technology can significantly reduce the intensity of artificial labor, which can be used to fisheries monitoring, aquaculture, etc. In recent years, the machine learning method based on image characteristics has been applied to fish image recognition, and it has achieved good results. However, ...

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): G06K9/46G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V10/44G06N3/045G06F18/241
Inventor 白静王艺然任俊杰牛林春
Owner XIDIAN 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