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

Freshwater fish identification method based on improved VGGNet

A recognition method and technology for freshwater fish, applied in neural learning methods, character and pattern recognition, fish farming, etc., and can solve problems such as water image deformation

Pending Publication Date: 2022-05-10
SHANDONG LANGCHAO YUNTOU INFORMATION TECH CO LTD
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] With the continuous development of related technologies such as machine learning and neural networks, classification methods relying on a large number of data sets and using machine learning libraries such as Tensorflow and Pytorch for training have emerged. When identifying, there are certain limitations due to certain deformation and other characteristics of the water image

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 identification method based on improved VGGNet
  • Freshwater fish identification method based on improved VGGNet
  • Freshwater fish identification method based on improved VGGNet

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0033] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, so that those skilled in the art can better understand the present invention and implement it, but the examples given are not intended to limit the present invention.

[0034] The present invention provides a freshwater fish recognition method based on improved VGGNet, which performs preprocessing and data enhancement for fish images in freshwater waters, and expands existing data sets.

[0035] Improve the network structure of VGGNet, select a small convolution kernel as the convolution kernel of the convolution layer, select the maximum pooling to keep the same pooling kernel parameters for each pooling layer, use three fully connected layers and then connect Softmax for classification ,

[0036] Using the improved VGGNet to train and verify the model according to the data set,

[0037] Freshwater fish identification by the model.

[0038] In...

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 identification method based on improved VGGNet, and relates to the technical field of fish identification. The method comprises the following steps: performing preprocessing and data enhancement on fish images in a freshwater area, expanding an existing data set, performing network structure improvement on VGGNet, selecting a small convolution kernel as a convolution kernel of a convolution layer, selecting a maximum pooling kernel parameter for keeping the same pooling kernel parameters of each pooling layer, performing classification by adopting a three-layer full-connection layer and then connecting Softmax, and performing classification by adopting a three-layer full-connection layer; and training and verifying a model according to the data set by using an improved VGGNet, and performing freshwater fish identification through the model.

Description

technical field [0001] The invention discloses a method and relates to the technical field of fish identification, in particular to an improved VGGNet-based freshwater fish identification method. Background technique [0002] With the continuous development of related technologies such as machine learning and neural networks, classification methods relying on a large number of data sets and using machine learning libraries such as Tensorflow and Pytorch for training have emerged. When identifying, there are certain limitations due to certain deformation and other characteristics of the water image. Contents of the invention [0003] Aiming at the problems of the prior art, the present invention provides a freshwater fish identification method based on the improved VGGNet, which optimizes the network structure and parameters of VGGNet, realizes fine-grained identification of freshwater fish with similar behavior, and has broad Application prospects. [0004] The concrete ...

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): G06V10/764G06V10/774G06V10/82G06V20/05G06K9/62G06N3/04G06N3/08A01K61/95
CPCG06N3/084A01K61/95G06N3/047G06N3/048G06N3/045G06F18/2415G06F18/214
Inventor 夏传涛李国涛胡清周永进
Owner SHANDONG LANGCHAO YUNTOU INFORMATION TECH CO LTD
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