Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Clustered adaptive window based hyperspectral image abnormality detection method

A hyperspectral image and anomaly detection technology, applied in image enhancement, image analysis, image data processing, etc., can solve problems such as hyperspectral image anomalies, achieve strong robustness and improve accuracy.

Active Publication Date: 2017-04-26
HARBIN INST OF TECH
View PDF8 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to solve the problem that the existing background model construction method is applicable when the hyperspectral image background is relatively consistent, and provides a hyperspectral image anomaly detection method based on clustering adaptive window

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
  • Clustered adaptive window based hyperspectral image abnormality detection method
  • Clustered adaptive window based hyperspectral image abnormality detection method
  • Clustered adaptive window based hyperspectral image abnormality detection method

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0013] Specific implementation mode one: the following combination figure 1 Describe this embodiment, the hyperspectral image anomaly detection method based on the cluster adaptive window described in this embodiment, the specific process of the hyperspectral image anomaly detection method is:

[0014] Step 1. Perform spectral dimension principal component analysis on hyperspectral images to generate spectral subspaces to achieve dimensionality reduction of spectral dimensions;

[0015] Step 2. Generate an adaptive window for each pixel to be detected; the generated adaptive window is a binary matrix, the center of the binary matrix coincides with the pixel to be detected, and a pixel with "1" in the matrix indicates that the hyperspectral image is the same The pixels in the homogeneous background area are used to calculate the statistical properties of the reference background. The pixels with "0" in the matrix indicate the pixels in the non-homogeneous background area in the...

specific Embodiment approach 2

[0019] Specific implementation mode two: the following combination figure 1 Describe this embodiment, this embodiment will further explain Embodiment 1, the specific process of performing spectral dimension principal component analysis on the hyperspectral image described in step 1 is:

[0020] Step 1-1, performing standardized transformation on the hyperspectral image matrix;

[0021] Step 1-2, obtaining the transformed correlation matrix;

[0022] Steps 1-3, obtaining the characteristic root of the correlation matrix, and determining the principal components of the hyperspectral image.

specific Embodiment approach 3

[0023] Specific implementation mode three: the following combination figure 1 Describe this embodiment, this embodiment will further explain Embodiment 1 or 2, the specific process of generating an adaptive window for each pixel to be detected as described in step 2 is:

[0024] Step 2-1. Generate a w×w adaptive window for each pixel to be tested, initialize the center of the adaptive window to "1", and initialize other areas to "0";

[0025] Step 2-2, using the Pearson criterion and the root mean square error minimization method to unmix the hyperspectral image, and determine the number of end members N;

[0026] Step 2-3, using the K-means clustering method to divide the hyperspectral image into N categories, and adding category labels to all pixels;

[0027] Steps 2-4: Compare the class labels of the pixels within the range of the adaptive window and the center point, and set the pixels in the adaptive window that are consistent with the label of the central pixel to "1" t...

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 provides a clustered adaptive window based hyperspectral image abnormality detection method which belongs to the hyperspectral image processing field with the object of solving the problem with the consistence of the hyperspectral image background restricted by an existing background model structuring method. The steps of the method are as follows: conducting analysis on the main components of spectral dimensions of hyperspectral image and generating spectral subspace; generating adaptive windows for each to-be-detected pixel wherein each of the generated adaptive window is a binary matrix whose center is superposed with the to-be-detected pixel and the pixel in the matrix represented by one indicates the pixel as one in the homogeneous background area of the hyperspectral image while the pixel in the matrix represented by zero indicates the pixel as one in the non-homogeneous background area of the hyperspectral image; using the analysis result of the main components and an elliptical contour model to estimate the background logarithmic likelihood of the adaptive window to detect the abnormal image elements and generate a preliminary matrix for detection result; and using the morphological filtering for post-treatment and obtaining the final result of the detection matrix. The invention is used to detect the abnormity of a hyperspectral remote sensing image.

Description

technical field [0001] The invention relates to abnormal detection of hyperspectral remote sensing images, belonging to the field of hyperspectral image processing. Background technique [0002] Limited by the spatial resolution of hyperspectral images, when the spectrum of the target to be detected is unknown or the target is small, it is easy to be submerged in the background of ground objects. At this time, the anomaly detection method is a very effective means for searching for such targets. Due to the lack of prior knowledge, the key process of anomaly detection is to estimate the background signal, and then highlight the difference between the target and the background and detect the target. Hyperspectral images have rich spectral information and complex spatial information. Improving the consistency of the reference background and more reasonable modeling can effectively improve the detection effect. Classical background model construction methods such as multivariat...

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
IPC IPC(8): G06T7/00
CPCG06T7/0002G06T2207/10036
Inventor 滕艺丹张钧萍江碧涛张晔时春雨钟圣唯
Owner HARBIN INST OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com