Unlock instant, AI-driven research and patent intelligence for your innovation.

Method for classifying hyperspectral images based on sparse characteristics and same neighborhood properties

A hyperspectral image and sparse feature technology, which is applied in hyperspectral image classification based on sparse features and neighborhood attributes, and in the field of hyperspectral image classification, can solve the problems of insufficient use of neighborhood information, long processing time, and low classification accuracy. Advanced problems, to achieve the effect of optimizing the classification effect, strong applicability, and reducing costs

Inactive Publication Date: 2014-07-02
HARBIN ENG UNIV
View PDF3 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] There are several problems in the traditional hyperspectral data classification method: 1. There are "noise" pixels in the hyperspectral data classification results
2. The classification accuracy is not high 3. The processing time is long
4. Not making full use of neighborhood information

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 classifying hyperspectral images based on sparse characteristics and same neighborhood properties
  • Method for classifying hyperspectral images based on sparse characteristics and same neighborhood properties
  • Method for classifying hyperspectral images based on sparse characteristics and same neighborhood properties

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0045] The present invention will be described in more detail below in conjunction with the accompanying drawings.

[0046] Specific steps are as follows:

[0047] 1. Read in the hyperspectral image data.

[0048] Read in the three-dimensional hyperspectral high-dimensional data, convert it from three-dimensional to two-dimensional data to facilitate subsequent processing, and normalize the obtained two-dimensional data to obtain X, and determine the data to be processed The number of sample categories is s.

[0049] 2. Solve the dictionary D.

[0050] Hyperspectral Remote Sensing Image Dataset dictionary (each column is an atom), the sparse representation can be expressed as an optimization problem of the following form:

[0051] min D , α 1 2 | | X - Dα ...

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 provides a method for classifying hyperspectral images based on the sparse characteristics and the same neighborhood properties. The method comprises the steps that 1, hyperspectral image data are read; 2, a dictionary D is solved; 3, the sparse characteristic A is solved; 4, a training set and a test set are set; 5, dichotomy is conducted on SVM; 6, a multi-classification result is determined; 7, a neighborhood dimension set C is read; 8, neighborhood division is conducted; 9, the same neighborhood properties are judged; 10, the step 8 and the step 9 are repeated and operated in a circulated mode until a same neighborhood property classification result Y1 is obtained and is a final classification result YM. The method has the advantages of being good in classification effect, low cost when high-dimensional data are processed, high in applicability and the like.

Description

technical field [0001] The invention relates to a hyperspectral image classification method, in particular to a hyperspectral image classification method based on sparse features and neighborhood homogeneity, and belongs to the technical field of remote sensing information processing. Background technique [0002] Hyperspectral data is characterized by a large amount of data, high redundancy, and high dimensionality. At the same time, there is a strong correlation between bands, which brings challenges to subsequent processing. Image object classification is the main content of remote sensing technology processing. Its basis is: the same type of pixel has consistency in spectral characteristics and spatial characteristics, and different object types have obvious differences in spectral characteristics and spatial characteristics. The remote sensing image classification process is the process of dividing pixels into the same category. The classification process occupies a ve...

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
Inventor 王立国杨京辉刘鲁涛林云
Owner HARBIN ENG UNIV