Supercharge Your Innovation With Domain-Expert AI Agents!

Hyperspectral Image Anomaly Detection Method Based on Non-negative Sparse Feature

A hyperspectral image, non-negative sparse technology, applied in the field of hyperspectral anomaly detection, can solve the problem of low accuracy of hyperspectral images, and achieve the effect of improving accuracy and accuracy

Active Publication Date: 2021-04-09
HARBIN INST OF TECH
View PDF10 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the problem of low accuracy of the existing hyperspectral image anomaly detection, and the present invention provides a hyperspectral image anomaly detection method based on non-negative sparse characteristics

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 Anomaly Detection Method Based on Non-negative Sparse Feature
  • Hyperspectral Image Anomaly Detection Method Based on Non-negative Sparse Feature
  • Hyperspectral Image Anomaly Detection Method Based on Non-negative Sparse Feature

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0047] Specific implementation mode one: see figure 1 Describe this embodiment, the hyperspectral image anomaly detection method based on the non-negative sparse characteristic described in this embodiment, the method includes the following steps:

[0048] Step 1: Obtain the non-negative sparse matrix S of the spectral image according to the hyperspectral image X t′ ;

[0049] Step 2: According to the non-negative sparse matrix S t′ Anomaly detection is performed on the hyperspectral image X, and the anomaly degree of each pixel in the hyperspectral image is obtained;

[0050] Step 3: Determine whether the pixel corresponding to each abnormality degree is abnormal according to the abnormality degree of each pixel in the hyperspectral image by means of threshold segmentation, so as to complete the abnormality detection of the hyperspectral image.

[0051] In this embodiment, in step 1, non-negative constraints are used to improve the sparse representation model of the hypers...

specific Embodiment approach 2

[0052] Specific implementation mode two: see figure 1 Describe this embodiment. The difference between this embodiment and the hyperspectral image anomaly detection method based on the non-negative sparse characteristic described in the first embodiment is that in the first step, the non-negative sparseness of the spectral image is obtained according to the hyperspectral image X Matrix S t′ The specific process is:

[0053] The hyperspectral image X is processed by the non-negative sparse coding method, and the non-negative sparse matrix S of the hyperspectral image is obtained. t′ .

[0054] In this embodiment, due to the addition of non-negative constraints, the sparse representation model of the hyperspectral image has obtained a more accurate and reasonable physical interpretation; Negative values ​​in the coefficient matrix will cause a certain degree of information loss. Using non-negative constraints can better solve this problem, and finally obtain a more accurate s...

specific Embodiment approach 3

[0055] Specific implementation mode three: see figure 1 This embodiment is described. The difference between this embodiment and the hyperspectral image anomaly detection method based on non-negative sparse characteristics described in Embodiment 1 or 2 is that in step 2, according to the non-negative sparse matrix S t′ Perform anomaly detection on the hyperspectral image X, and the specific process of obtaining the anomaly degree of each pixel in the hyperspectral image is as follows:

[0056] Non-negative sparse matrix S for hyperspectral images t′ Sparse score estimation is performed to calculate the abnormality of each pixel in the hyperspectral image.

[0057] In this embodiment, the present invention combines the hyperspectral image non-negative sparse coding theory and the sparse score estimation algorithm, and proposes a hyperspectral image anomaly detection method based on the non-negative sparse characteristic. In this process, the sparse nature of the anomaly is f...

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

A hyperspectral image anomaly detection method based on a non-negative sparse characteristic belongs to the field of hyperspectral anomaly detection. Solve the problem of low accuracy of existing hyperspectral image anomaly detection. The method of the present invention comprises the following steps: Step 1: Obtain the non-negative sparse matrix S of the spectral image according to the hyperspectral image X t ’; Step 2: According to the non-negative sparse matrix S t 'Carry out anomaly detection on the hyperspectral image X to obtain the anomaly degree of each pixel in the hyperspectral image; Step 3: Through threshold segmentation, determine the corresponding Whether the pixel is abnormal, so as to complete the abnormal detection of hyperspectral image. The invention is mainly used for anomaly detection on hyperspectral images.

Description

technical field [0001] The invention belongs to the field of hyperspectral anomaly detection. Background technique [0002] Hyperspectral remote sensing is an information acquisition method that can conduct fine observation of ground objects. The abnormalities in hyperspectral images have the characteristics of low probability of occurrence and small number of existence, and the amount of information contained is relatively rich, so it has certain research value. Due to the sparse nature of the anomaly itself, sparse representation has a good application prospect in the field of hyperspectral anomaly detection. [0003] However, how to effectively use sparse representation for anomaly detection is still in the exploration stage. In addition, the current sparse coefficient solution does not consider the non-negative constraints of the coefficients, resulting in negative values ​​in the coefficients, which violates the non-negative constraints of the hyperspectral linear spec...

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): G06T7/10G06T7/136G06K9/62
CPCG06T7/10G06T7/136G06T2207/10036G06F18/28
Inventor 张钧萍孙邱鹏
Owner HARBIN INST OF TECH
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com