Hyperspectral image classification multi-kernel learning method capable of maximizing class separability

A hyperspectral image, multi-core learning technology, applied in character and pattern recognition, instruments, computer parts and other directions, can solve the problem of low solution efficiency, without considering multi-core model solution and subsequent classification applications, etc., to achieve high algorithm efficiency, improve The efficiency of algorithm operation and the effect of improving classification performance

Active Publication Date: 2017-05-31
HARBIN INST OF TECH
View PDF2 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In order to solve the problem that the existing multi-kernel learning method applied to hyperspectral image classification does not consider the combination of multi-kernel model solving and subsequent classification applications and the problem of low solution efficiency, the present invention proposes a criterion based on maximizing category separability multi-core learning method

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
  • Hyperspectral image classification multi-kernel learning method capable of maximizing class separability
  • Hyperspectral image classification multi-kernel learning method capable of maximizing class separability
  • Hyperspectral image classification multi-kernel learning method capable of maximizing class separability

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0017] Specific implementation mode 1. Combination figure 1 To illustrate this embodiment, a multi-core learning method based on maximizing category separability criterion described in this embodiment is performed in the following steps:

[0018] Step 1. Obtain training samples and test samples from a given input hyperspectral image dataset;

[0019] Step 2. Use the training sample set Construct the base kernel matrix K in the multi-kernel learning model m , get the base kernel matrix set

[0020] Step 3, using the base kernel matrix set to measure the intra-class dispersion and inter-class dispersion of the hyperspectral image dataset in the Hilbert kernel space;

[0021] Step 4: On the basis of intra-class dispersion and inter-class dispersion, class separability is measured with the maximum class interval criterion, and the maximum class separability is used as the solution criterion for the multi-kernel learning model to solve the base kernel weight.

[0022] This e...

specific Embodiment approach 2

[0025] Specific embodiment 2: The difference between this embodiment and specific embodiment 1 is that in the step 1, training samples and test samples are obtained from a given input hyperspectral image data set; the specific process is:

[0026] Step 11. The given input hyperspectral image dataset is: x i ∈R F

[0027] Among them, T represents the total number of samples in the input hyperspectral image dataset, Input hyperspectral image dataset X input Contains samples of class C, and the number of samples of class l is N l ;1≤l≤C;x i For the i-th sample in the input hyperspectral image dataset, R F is a space representing dimension F, F is the feature dimension of the input hyperspectral image data set, and i is a positive integer;

[0028] Step 12. From the input hyperspectral image dataset X input Randomly select samples according to p% of each class to form a training sample set:

[0029] Among them, p is determined by the complexity of the input data and s...

specific Embodiment approach 3

[0031] Specific embodiment three: the difference between this embodiment and specific embodiment one or two is that the training sample set is used in the second step Construct the base kernel matrix K in the multi-kernel learning model m , get the base kernel matrix set The base kernel matrix is ​​constructed as follows:

[0032] The training sample set The middle samples are input in pairs to the Gaussian kernel function to calculate the base kernel matrix where σ m is the Gaussian kernel scale parameter, x j is the jth sample in the training sample set, and i and j are positive integers.

[0033] Other steps and parameters are the same as those in Embodiment 1 or Embodiment 2.

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 hyperspectral image classification multi-kernel learning method capable of maximizing the class separability, and relates to multi-kernel learning model solving. The invention aims to solve problems that multi-kernel model solving is not considered to be combined with subsequent classification application and the solving efficiency is low in an existing multi-kernel learning method applied to hyperspectral image classification. The hyperspectral image classification multi-kernel learning method is implemented according to the following steps: step one, training samples and test samples are acquired from a given input hyperspectral image data set; step two, a base kernel matrix Km in the multi-kernel learning model is constructed by using the training sample set Xtrain={xi}<i=1><N>, and a base kernel matrix set {Km}<m=1><M>={K1, K2,..., Km} is acquired; step three, the within-class discrete degree and the between-class discrete degree of the data set in a Hilbert kernel space are measured by using the base kernel matrix set; and step four, the class separability is measured according to a maximum class interval principle on the basis of the within-class discrete degree and the between-class discrete degree, and the weight of the base kernel is solved by taking the maximum class separability as a solving principle of the multi-kernel learning model. The hyperspectral image classification multi-kernel learning method is applied to the field of pattern recognition.

Description

technical field [0001] The invention belongs to the field of pattern recognition and relates to solving multi-core learning models. Background technique [0002] Existing multi-kernel learning methods can be summarized into three categories. The first type is a multi-core learning method based on fixed criteria. This type of method uses fixed criteria such as addition and multiplication to combine base cores during the multi-core learning process, without considering the similarity measurement characteristics of each base core and without using any optimization criteria. Hence the classification performance is lower. The second type is the single-step optimization multi-core learning method. This type of method solves the weight parameters of the base core in the multi-core model and the parameters in the classifier at the same time, so the calculation amount is very large, the learning efficiency is very low, and its classification performance is relatively strong. The th...

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): G06K9/62G06K9/46
CPCG06V10/462G06F18/2411
Inventor 谷延锋王青旺
Owner HARBIN INST 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