A plant species identification method based on leaf and flower fusion for locally distinguishing CCA

A recognition method and species technology, applied in the field of image recognition, can solve the problem of not considering the relationship of multi-modal data sets, and achieve the effect of accurate recognition, stable performance, and retention of identification information and data structure.

Pending Publication Date: 2019-06-28
TIANJIN UNIVERSITY OF SCIENCE AND TECHNOLOGY
View PDF3 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, they are single-view learning methods that do not consider the r

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
  • A plant species identification method based on leaf and flower fusion for locally distinguishing CCA
  • A plant species identification method based on leaf and flower fusion for locally distinguishing CCA
  • A plant species identification method based on leaf and flower fusion for locally distinguishing CCA

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0037] Embodiments of the present invention will be described in further detail below in conjunction with the accompanying drawings.

[0038] The design concept of the present invention is: the present invention introduces the idea of ​​local discriminant embedding (LDE) into canonical correlation analysis (CCA), and constitutes local discriminant canonical correlation analysis (MLDCCA) method for plant species identification, which by maximizing the Correlation of neighboring samples and minimizing the correlation of neighboring samples from different classes, then two projection matrices are obtained to achieve feature extraction. Its main technical means are: (1) Using label information and data structure to combine multiple organs to identify plant species. (2) Two neighbor graphs are constructed to measure inter-class separability and intra-class compactness in a local manner. (3) Introduce geodesic distance into 1-NN classifier.

[0039] Firstly, the relevant technolog...

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 plant species identification method based on leaf and flower fusion for local discrimination of CCA, comprising the following steps: constructing two weighted adjacent graphs according to a double-view feature set comprising a leaf image and a flower image; Obtaining two dimensionality-reduced projection matrixes by enabling the adjacent samples in the class to be most correlated, enabling the adjacent samples between the classes to be most uncorrelated and enabling the correlation between leaves and flowers of the same species to be maximum; identifying plant species using a 1-NN classifier with a geodesic distance. According to the method, the idea of local discrimination embedding (LDE) is introduced into canonical correlation analysis (CCA); The method has the advantages that the identification is accurate, the performance is stable, the identification characteristics can be extracted from the two plant organs, the identification information and the datastructure can be well reserved, and the method is further popularized and applied on the basis.

Description

technical field [0001] The invention belongs to the technical field of image recognition, in particular to a plant species recognition method based on leaf-flower fusion and local discrimination CCA. Background technique [0002] Automatic plant species identification is crucial to the conservation of species diversity, which can help ordinary people and botanists identify various plant species more quickly. Plant species can be identified by their leaves, flowers, bark, branches, seeds, fruits, and whole bodies, and leaf-based automatic plant species identification is an important area in several fields such as computer science, image processing, algorithm science, botany, and machine learning. important topic. By extracting various features from leaves and using them for plant species recognition, the recognition rate largely depends on manual features extracted from plant leaf images. Deep learning can automatically learn classification features from original images, wh...

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): G06K9/62
Inventor 张传雷刘丽欣李建荣武大硕任雪飞刘璞张善文
Owner TIANJIN UNIVERSITY OF SCIENCE AND TECHNOLOGY
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