Hyperspectral data noise reduction method and system based on spatial correlation
A spatial correlation and data noise reduction technology, applied in the field of noise reduction of hyperspectral data, can solve problems such as high signal-to-noise ratio and lack of reach
Inactive Publication Date: 2017-08-25
WUHAN UNIV
View PDF1 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Therefore, the noise reduction effect has certain limitations and cannot achieve a higher signal-to-noise ratio.
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 Construction
[0069] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.
[0070] During specific implementation, the present invention can use computer software technology to realize the automatic operation process. For the technical solution of the present invention in detail, refer to figure 1 , the specific description of the embodiment process is provided as follows:
[0071] Such as figure 1 As shown, the embodiment of the present invention sequentially includes a projection transformation step, a noise reduction threshold estimation step, a noise reduction step, and a projection inverse transformation step. The embodiment of the present invention adopts the hyperspectral data collected in Indiana by AVIRIS. The data size is 145×145×220, that is, a total of 220 bands. The images of each band contain 145×145 pixels. Due to the interference of water vapor and instruments, Part of the band (150-163) was removed. figure...
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 provides a hyperspectral data denoising method and system based on spatial correlation, which performs projection transformation, solves the average image of images formed by each band in hyperspectral data, calculates the covariance matrix of hyperspectral data and performs eigenvalue Decompose to obtain the transformation matrix and eigenvalue matrix, use the transformation matrix to linearly project the hyperspectral data to obtain the three-dimensional data in the transformation domain; select the noise reduction threshold according to the eigenvalue matrix; noise reduction, including the transformation domain within the noise reduction threshold The three-dimensional data in the transformation domain is retained, and the three-dimensional data in the transformation domain beyond the threshold is subjected to noise reduction processing; the inverse projection transformation includes using the inverse matrix of the transformation matrix to linearly project the three-dimensional data in the transformation domain after noise reduction, and re The noise-reduced hyperspectral data is constructed. During the noise reduction process, the present invention fully protects the correlation of the space domain, well reduces the noise level of the hyperspectral data, and improves the signal-to-noise ratio of the hyperspectral data.
Description
technical field [0001] The invention belongs to the technical field of hyperspectral data processing, and in particular relates to a hyperspectral data noise reduction method, which is used to reduce the noise level of hyperspectral data collected by sensors and improve the quality of hyperspectral data. Background technique [0002] Hyperspectral data is a multidimensional signal obtained using hyperspectral remote sensing technology, including two-dimensional image data and one-dimensional spectral data of the target area. Hyperspectral data generally includes hundreds of spectral information, and the wavelength range includes ultraviolet, visible light, infrared, etc., so at each sampling point, continuous spectral information is formed. In recent years, hyperspectral data has been applied in fields such as target recognition, agricultural and forestry monitoring, and map filling. However, in the process of acquisition and transmission, the hyperspectral image is subject...
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 Patents(China)
IPC IPC(8): G06T5/00
Inventor 马佳义马泳黄珺
Owner WUHAN UNIV