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

Hyperspectral Optimal Band Selection Method Based on Shared Nearest Neighbor

An optimal band and hyperspectral technology, applied in the field of image processing, can solve problems such as poor practicability, and achieve the effect of good practicability and improved robustness

Active Publication Date: 2022-04-29
NORTHWESTERN POLYTECHNICAL UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In order to overcome the shortcomings of poor practicability of existing band selection methods, the present invention provides a hyperspectral optimal band selection method based on shared nearest neighbors

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 Optimal Band Selection Method Based on Shared Nearest Neighbor
  • Hyperspectral Optimal Band Selection Method Based on Shared Nearest Neighbor
  • Hyperspectral Optimal Band Selection Method Based on Shared Nearest Neighbor

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0028] refer to figure 1 . The specific steps of the hyperspectral optimal band selection method based on shared nearest neighbor in the present invention are as follows:

[0029] Step 1. Assuming that L is the number of hyperspectral bands, and the size of each band is W×H, the spatial image of each band is stretched into a one-dimensional vector, and the data of each band is normalized. Get the initial matrix X=[x for all bands 1 ,x 2 ,...,x L ], where x i is the vector of band i.

[0030] Step 2. Use Euclidean distance to measure the distance between any two bands as D(x i ,x j ):

[0031]

[0032] Using the K nearest neighbor method to obtain the K bands around the band i are:

[0033] kd(x i )={x i ∈X|D(x i ,x j )≤d i},

[0034] Among them, d i Indicates the distance to the Kth band from band i.

[0035] Step 3. For each band of the hyperspectrum, calculate the number of local shared neighbors:

[0036] S(x i ,x j )=|kd(x i )∩kd(x j )|,

[0037] ...

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 optimal band selection method based on shared nearest neighbors, which is used to solve the technical problem of poor practicability of the existing band selection methods. The technical solution is to use the Euclidean distance to measure the similarity between each band, obtain K adjacent bands around each band through the K nearest neighbor method, and use the shared nearest neighbor method to calculate the local density of each band; obtain each band to other The minimum distance of the high-density band, calculate the amount of information in each band through information entropy, and use the product of the three factors as the band weight; sort the weights of the hyperspectral band in descending order, and obtain the maximum index through the slope change of the weight curve, and then Determine the optimal number of bands. Since the shared nearest neighbor locally analyzes the local similarity between each band and other bands, it can accurately reflect the local distribution characteristics of each band in space, and at the same time, considering the amount of information of the selected band, it improves the robustness of hyperspectral band selection. Rod, good practicality.

Description

technical field [0001] The invention belongs to the technical field of image processing, in particular to a hyperspectral optimal band selection method based on shared nearest neighbors. Background technique [0002] Using hyperspectral sensors, a large number of continuous band images are captured through different wavelengths. Compared with RGB images, these bands can provide richer spectral information and image information, and can continuously image the same target object to better describe the ground. The difference in spectral characteristics of objects improves the detection and identification capabilities of targets. However, a large number of hyperspectral bands doubles the amount of data, and the information redundancy between bands is high, which is not conducive to subsequent image analysis. Therefore, it is very necessary to reduce the amount of data and save resources. In order not to change the original data, the band selection technology is applied, which ...

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 Patents(China)
IPC IPC(8): G06K9/00G06K9/62
Inventor 王琦李学龙李强
Owner NORTHWESTERN POLYTECHNICAL UNIV