Bird endangered species identification method based on structure-preserving zero-sample learning

A technology for endangered species and identification methods, applied in character and pattern recognition, computer parts, instruments, etc., can solve the problems of reduced accuracy, achieve classification accuracy, solve domain drift, and retain structural information.

Active Publication Date: 2020-01-21
SOUTH CHINA UNIV OF TECH
View PDF3 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In this case, directly applying the mapping learned from the seen category data to the unseen category classification task will lead to a decrease in accuracy

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
  • Bird endangered species identification method based on structure-preserving zero-sample learning
  • Bird endangered species identification method based on structure-preserving zero-sample learning
  • Bird endangered species identification method based on structure-preserving zero-sample learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0037] This example proposes a method for identifying endangered bird species based on structure-preserving zero-shot learning, aiming to overcome the problems existing in existing methods and solve the problem of identifying endangered bird species. The present invention can realize identification of endangered bird species by using common bird image features for which label information can be obtained and descriptive information of endangered bird species when the image information of endangered birds cannot be obtained. Such as figure 1 Shown is a schematic flow chart of the identification method of the present invention, according to figure 1 , the corresponding specific implementation steps are as follows:

[0038] (1) Data input steps

[0039] The image data that needs to be input is divided into visible category data and invisible category data. The visible categories refer to common birds for which label information is available, and the invisible categories refer t...

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 bird endangered species identification method based on structure-preserving zero sample learning, and the method comprises the following steps: a data input step: inputting adata set which comprises the visual features, semantic information and label information of a common bird image, and the semantic information of bird endangered species; and a training step: learningbidirectional mapping from a visual feature space to a semantic space, and meanwhile, and further constraining the mapping by using manifold consistency, wherein the optimization problem is summarized into a Sylvester equation solving problem, the solving process is simple and easy to implement, and the solving result is a mapping matrix P; and a prediction step of identifying the bird endangeredspecies image with the given semantic information by using the mapping matrix P obtained in the training step. According to the method, structural information between data is reserved, the problem ofdomain drift is solved, the image classification accuracy is improved, the method can be applied to complex bird image recognition, and endangered species without known label information can be recognized.

Description

technical field [0001] The invention relates to the technical field of image classification, in particular to a method for identifying endangered species of birds based on structure-preserving zero-sample learning. Background technique [0002] Birds are an important part of the animal kingdom. Due to their wide variety and similar appearance, how to identify them accurately and efficiently has always been an important research problem. The traditional field of computer vision requires a large number of human-labeled bird images for training to obtain good classification accuracy. However, due to changes in the natural environment and the impact of human activities, the number of many birds has declined sharply, and relevant images have become difficult to obtain. Information about their appearance and habits can only be obtained from ancient books and descriptions by insiders. Due to the scarcity of labeled image data, methods in the traditional computer vision field canno...

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): G06K9/62
CPCG06F18/24143
Inventor 周智恒牛畅尚俊媛黄俊楚张鹏宇
Owner SOUTH CHINA UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap