Hyperspectral image extreme learning machine clustering method based on space-spectrum joint hypergraph embedding
A hyperspectral image and extreme learning machine technology, which is applied in the field of hyperspectral image extreme learning machine clustering embedded in space-spectrum joint hypergraph, can solve the problems of no effective joint space-spectral information and low clustering accuracy
Active Publication Date: 2019-10-22
HENAN UNIVERSITY
View PDF4 Cites 8 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] Aiming at the problem that the existing hyperspectral image clustering algorithm does not effectively combine space-spectral information and the clustering accuracy is low, the present invention proposes a hyperspectral image extreme learning machine clustering method based on space-spectrum joint hypergraph embedding
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 2
[0080]Another hyperspectral image extreme learning machine clustering method based on joint hypergraph embedding of space and spectrum, including:
[0081] Step S201: hyperspectral data preprocessing, including:
[0082] Input hyperspectral image X 0 ∈R D×W×H , D, W, H respectively represent the feature dimension of the hyperspectral image and the width and height of the spatial dimension, as a possible implementation, with image 3 The PaviaUniversity data set image shown in part (a) is an experimental example, correspondingly, D=103, W=100, H=200, and the hyperspectral data X 0 Arranged row by pixel to form preprocessed hyperspectral data As the input of the extreme learning machine model, where N=W×H represents the number of hyperspectral pixels, x i ∈R D Represents a hyperspectral pixel.
[0083] Step S202: Calculate the space-spectrum joint information of the preprocessed hyperspectral data, including:
[0084] Step S202.1: for (m i ,n i ) hyperspectral pixel x ...
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
Login to View More Abstract
The invention belongs to the technical field of image processing, and discloses a hyperspectral image extreme learning machine clustering method based on space-spectrum joint hypergraph embedding, which comprises the following steps: step 1, carrying out hyperspectral data preprocessing; 2, calculating space-spectrum joint information XS of the preprocessed hyperspectral data; 3, constructing a space-spectrum joint hypergraph through XS; 4, calculating a hypergraph Laplace matrix Lh through the space spectrum combined hypergraph; 5, setting a hidden layer network weight parameter; 6, calculating hidden layer features; 7, constructing a space-spectrum combined hypergraph Laplace regular term and an optimization model; and step 8, solving the optimization model to obtain similarity preserving projection characteristics of the space-spectrum combined hypergraph structure, and performing spectral clustering to obtain a final clustering label. The method is high in clustering precision andhigh in noise robustness.
Description
technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a hyperspectral image extreme learning machine clustering method embedded in a space-spectrum joint hypergraph. Background technique [0002] Hyperspectral remote sensing is a technology that uses imaging spectrometers to obtain tens to hundreds of very narrow (usually 2-10nm) and spectrally continuous image data in the visible, near-infrared, mid-infrared and thermal infrared bands of the electromagnetic spectrum. Hyperspectral remote sensing images are widely used in deep space exploration, earth observation and quantitative remote sensing due to their rich triple information of space, radiation and spectrum. At present, the problem of hyperspectral image clustering has been widely concerned by scholars and has achieved good applications. The basic principle of hyperspectral image clustering is to assign similar pixels to the same category on the basis of ...
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
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/62G06K9/00
CPCG06V20/13G06V20/194G06F18/23
Inventor 夏浩铭秦耀辰陈优阔赵威
Owner HENAN UNIVERSITY