Hyperspectral image classification method based on area similarity low rank expression dimension reduction

A hyperspectral image and low-rank representation technology, applied in the field of remote sensing image processing, can solve the problems of large data volume, loss of data global information, and ineffective use of hyperspectral image space information, so as to reduce computational complexity and improve accuracy The effect of reducing the amount of calculation

Active Publication Date: 2013-12-04
XIDIAN UNIV
View PDF5 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, due to the large amount of hyperspectral data, it is impossible to construct a similarity matrix in manifold learning such as ISOMAP, which limits the application of many manifold learning dimensionality reduction methods on hyperspectral images, and loses the global information of the data.
[0004] The main problems of the current hyperspectral image dimensionality reduction methods are: only using the spectral information of the hyperspectral image, and not effectively utilizing the spatial information of the hyperspectral image; the amount of data to be processed is large, and the image after dimensionality reduction cannot be very Good preservation of all useful information provided by the original hyperspectral image data

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 method based on area similarity low rank expression dimension reduction
  • Hyperspectral image classification method based on area similarity low rank expression dimension reduction
  • Hyperspectral image classification method based on area similarity low rank expression dimension reduction

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0036] The invention proposes a hyperspectral image classification method based on low-rank representation of region similarity and dimensionality reduction. At present, hyperspectral images are extremely important in military and civilian applications. However, because the rich spectral information of hyperspectral images not only contains a large amount of redundant information, but also the huge amount of data also affects the classification efficiency and classification accuracy of hyperspectral images, so the dimensionality reduction of hyperspectral images is important in hyperspectral image classification. It plays a very important role in the application. In view of the fact that the existing hyperspectral image dimensionality reduction methods do not fully utilize the spatial structure information of hyperspectral images and all the useful information provided by hyperspectral images, combined with mean shift pre-segmentation and low-rank representation, the present i...

Embodiment 2

[0059] Hyperspectral image classification method based on regional similarity low-rank representation dimensionality reduction, same as embodiment 1

[0060] 1. Simulation conditions:

[0061] The simulation experiment uses the Indian Pine image acquired by the airborne visible light / infrared imaging spectrometer AVIRIS of NASA Jet Propulsion Laboratory in June 1992 in northwestern Indiana, such as image 3 As shown, the image size is 145×145, with a total of 220 bands, and there are 200 bands for removing noise and atmospheric and water absorption bands, with a total of 16 types of ground object information.

[0062] The simulation experiment is carried out with MATLAB R2012b software on the WINDOWS7 system with Intel Core(TM) 2Duo CPU, main frequency 2.33GHz, and 2G memory.

[0063] 2. Simulation content and analysis:

[0064] Using the present invention and the existing four methods to classify the hyperspectral image Indian Pine, the existing four methods are: support vect...

Embodiment 3

[0069] The hyperspectral image classification method based on low-rank representation of regional similarity and dimensionality reduction is the same as that in Embodiment 1 and Embodiment 2

[0070] Select some pixels from the Indiana Pine data as marked pixels, and the remaining pixels as unmarked pixels. The present invention and the existing four methods carry out 30 classification experiments on the Indiana Pine data, and get the average value of the classification results as The final classification accuracy rate, as shown in Figure 4, Figure 4b It is a graph of the relationship between the classification accuracy rate and the feature dimension of the five methods when the marked samples are 10% of the total number of samples. The abscissa is the feature dimension, and the dimension range is 1 to 30. The ordinate is the classification accuracy rate. . From Figure 4b It can be seen that when the feature dimension is greater than 12, the classification accuracy rate of...

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 hyperspectral image classification method based on area similarity low rank expression dimension reduction. The method includes first expressing each pixel point of a hyperspectral image through characteristic vectors, conducting area pre-segmentation by using mean shifting, utilizing each image block mean obtained in pre-segmentation as a new sample to build a new sample set, learning a low rank coefficient matrix of the new sample set, building a segmented image area similarity matrix, building a characteristic value equation, calculating the equation to obtain a projection matrix, projecting hyperspectral data to the low-dimensional space to obtain a sample set subjected to dimension reduction and inputting the sample set subjected to the dimension reduction into a support vector machine to conduct classification and obtain a classification label. By means of the method, the problems of large calculation quantity and low accuracy of the existing method and caused by high data dimension of the hyperspectral image are solved, the space structure information of hyperspectral image is utilized, overall information of the hyperspectral image is effectively expressed, hyperspectral image classification accuracy is improved, and the method can be used in the civil field and the military fields of map making, vegetation investigation, military intelligence acquisition and the like.

Description

technical field [0001] The invention belongs to the technical field of remote sensing image processing, and relates to a low-rank representation method based on regional similarity, specifically a hyperspectral image classification method based on low-rank representation of regional similarity for dimensionality reduction, which is used for data dimensionality reduction and classification of hyperspectral remote sensing images . Background technique [0002] Hyperspectral remote sensing technology uses imaging spectrometers to image surface objects to form three-dimensional data composed of continuous band images. It has been widely used in target recognition, precision agriculture, resource exploration, environmental monitoring and other fields. The images acquired by hyperspectral remote sensing technology contain rich spatial, radiation and spectral triple information. On the one hand, a large amount of information provides a large amount of basis for pixel classification...

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 & AuthorityApplications(China)
IPC IPC(8): G06K9/62G06K9/46
Inventor张向荣焦李成贺予迪侯彪王爽吴家骥马文萍马晶晶
OwnerXIDIAN UNIV