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

Method for identifying Pelteobagrus fulvidraco and its interbreeding specie based on depth convolution generation antagonistic network

A technology of deep convolution and recognition methods, applied in biological neural network models, character and pattern recognition, instruments, etc., can solve the problems of low recognition accuracy, laborious, time-consuming data preprocessing, etc., achieve good recognition effect, improve general The effect of chemicalization ability and high-precision production automation

Inactive Publication Date: 2019-01-01
NANJING NORMAL UNIVERSITY
View PDF5 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Purpose of the invention: Aiming at the deficiencies of the prior art, the present invention provides a method for identifying yellow catfish and its nested species based on deep convolutional generative adversarial networks, combining deep learning methods and fish identification methods, aiming at Solve the shortcomings of low identification accuracy, time-consuming and laborious data preprocessing of traditional yellow catfish and its nested species

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
  • Method for identifying Pelteobagrus fulvidraco and its interbreeding specie based on depth convolution generation antagonistic network
  • Method for identifying Pelteobagrus fulvidraco and its interbreeding specie based on depth convolution generation antagonistic network
  • Method for identifying Pelteobagrus fulvidraco and its interbreeding specie based on depth convolution generation antagonistic network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0033] The technical solution of the present invention will be further described below in conjunction with the accompanying drawings.

[0034] refer to figure 1 , the method of the present invention is by collecting the original picture of yellow catfish and its nested fish, and generates a realistic picture similar to the original picture through a generative adversarial network (Generative adversarial networks, GAN); Mixed and sent to the deep convolutional neural network for training; in practical application, the trained deep convolutional neural network is used to identify the yellow catfish and its nested species of fish. The invention generates new samples through a generative confrontation network, and increases the number of samples through affine transformation (rotation, stretching, shearing, etc.), increases the richness of samples, and improves the generalization ability of the model.

[0035] In one embodiment, according to the nesting mode of yellow catfish, th...

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 Pelteobagrus fulvidraco and an interbreeding species identification method thereof based on a depth convolution generation antagonistic network, belonging to the technical field of machine learning. The method comprises the following steps of: collecting original pictures of yellow catfish and interbreeding fish of yellow catfish, and generating fish pictures similar to the original pictures through a generated antagonistic network; the fish images are mixed with the original images and sent to the convolution neural network for training. The trained convolution neural network was used to identify Pelteobagrus fulvidraco and its interbreeds. The method can generate images similar to the real Pelteobagrus fulvidraco and its interbreeding species under the conditionof very few samples, and thoroughly solves the problem that Pelteobagrus fulvidraco living in water is hard to collect images out of water, and has hard spines on the body surface, which leads to manual sorting and easy injury. By using the original data and the pictures generated by the depth convolution countermeasure network for the input of the convolution neural network, the recognition accuracy of the network model is significantly improved, and the final recognition accuracy can reach 94.2%, which has important application value.

Description

technical field [0001] The invention relates to an intelligent aquaculture method, in particular to a method for identifying yellow catfish and its nested species based on a deep convolution generation confrontation network. Background technique [0002] Yellow catfish (Pseudobagrus fulvidraco) is the main small economic fish in agricultural aquaculture in my country. As the yellow catfish is an omnivorous fish, with the expansion of the yellow catfish industrial system, it is beneficial to make full use of the feeding habits of different fish, make full use of the bait, purify the water quality, and improve the pond feed. The conversion rate is conducive to increasing production and is beneficial to fishermen in earning foreign exchange and competing for income. The interbreeding mode of yellow catfish mainly includes: filter-feeding fish and yellow catfish interbreeding, filter-feeding farmed fish mainly include silver carp, bighead carp, etc.; herbivorous fish and yellow...

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): G06K9/62G06N3/04G06N3/06
CPCG06N3/061G06N3/045G06F18/214
Inventor 尹绍武谢万里王涛张红燕
Owner NANJING NORMAL UNIVERSITY
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