High spectral image object detection method based on tensor principal component analysis dimension reduction

A technology of hyperspectral image and principal component analysis, which is applied in the field of hyperspectral image target detection, can solve the problems of low detection accuracy and inability to mine the overall information of 3D data, and achieve the effect of improving detection accuracy and effective detection

Active Publication Date: 2017-10-10
HARBIN INST OF TECH
View PDF8 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to solve the problem that the existing hyperspectral image target detection method does not fully consider the characteristics of spatial constraint enhancement under high-resolution conditions, cannot

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
  • High spectral image object detection method based on tensor principal component analysis dimension reduction
  • High spectral image object detection method based on tensor principal component analysis dimension reduction
  • High spectral image object detection method based on tensor principal component analysis dimension reduction

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0030] Specific implementation mode one: combine figure 1 Describe this embodiment, a hyperspectral image target detection method based on tensor principal component analysis dimensionality reduction in this embodiment, the specific process is:

[0031] Step 1: Select and divide the tensor block of the hyperspectral image to be detected, and obtain the tensor block of the empty X-empty Y-spectrum third-order target template, the tensor block of the third-order background template of the empty X-empty Y-spectrum and the tensor block to be Detected empty X-null Y-spectrum third-order test sample tensor block;

[0032] Step 2: Set the size of the target template tensor block, background template tensor block and test sample tensor block to be tested in each dimension after projection, and use all test sample tensor blocks to obtain the target template tensor block , the projection matrix in three dimensions of the background template tensor block and the test sample tensor bloc...

specific Embodiment approach 2

[0036] Specific embodiment 2: The difference between this embodiment and specific embodiment 1 is that in the first step, the hyperspectral image to be detected is selected and divided into tensor blocks, and the empty X-empty Y-spectrum third-order target template sheet is obtained. Gauge block, empty X-empty Y-spectrum third-order background template tensor block and empty X-empty Y-spectrum third-order test sample tensor block to be detected; the specific process is:

[0037] Given a 3×3 window, convert the hyperspectral image to be detected into a third-order tensor form, slide the sampling window, and determine it as empty X-empty when the value of the truth map corresponding to the center point of the sampling window is 1 Y-spectrum third-order target template tensor block T-tensor(i), i=1,2,...N T ; When the values ​​of all points in the window corresponding to the truth map are 0, it is determined to be empty X-empty Y-spectrum third-order background template tensor bl...

specific Embodiment approach 3

[0039] Specific embodiment three: the difference between this embodiment and specific embodiment one or two is: in the step two, set the target template tensor block, the background template tensor block and each dimension projection of the test sample tensor block to be detected After the size of the dimension, use all the test sample tensor blocks to be detected to obtain the projection matrix in the three dimensions of the target template tensor block, the background template tensor block and the test sample tensor block to be detected; the specific process is:

[0040] Step 21. Set the size P of the projected dimensions of the target template tensor block, the background template tensor block, and the test sample tensor block to be tested. n ,n=1,2,3;

[0041] Step 22: Using the tensor principal component analysis algorithm to use all the test sample tensor blocks to be detected to obtain the projection matrix in the three dimensions of the target template tensor block, th...

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 high spectral image object detection method based on tensor principal constituent analysis dimension reduction is provided; an existing high spectral image object detection method cannot fully consider enhanced space constraint characteristics under high resolution conditions, cannot excavate information from the three dimensional data integral body, and is low in detection precision; the high spectral image object detection method based on tensor principal constituent analysis dimension reduction can solve said problems, and comprises the following steps: 1, obtaining third order object, third order background and detected third order test sample tensor blocks; 2, obtaining a projection matrix on three dimensions including the object, background and detected test sample; 3, projecting the object, background and detected test sample to a preset tensor subspace; 4, calculating the total distance from each detected test sample to the background and an object template; 5, using the distance ratio as a gray value, and determining the central point pixel as the object if the gray value is bigger than a threshold value, otherwise determining the central point pixel as the background. The method is applied to the image processing field.

Description

technical field [0001] The invention relates to a hyperspectral image target detection method. Background technique [0002] The hyperspectral image sensor can obtain the reflected radiation information of ground objects through hundreds of spectral channels, and its band range covers from visible light to near infrared and even long wave infrared. Hyperspectral images contain spatial information, reflection or radiation information, and spectral information of ground objects at the same time, and its characteristic is usually called "map-spectrum integration". Moreover, hyperspectral image data provides nearly continuous spectral sampling information, which can record small reflection differences of ground objects in the spectrum, which can be used as the basis for classification and detection of ground objects. It has important theoretical significance and application value to study the new technology of hyperspectral image target detection. In the military aspect, it ca...

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): G06K9/00G06K9/62
CPCG06V20/13G06V20/194G06F18/2135G06F18/24
Inventor 谷延锋谭苏灵
Owner HARBIN INST OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap