Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Real-time segmentation and classification method of passive terahertz image target

A classification method and terahertz technology, applied in the field of terahertz image processing, can solve problems such as inability to accurately cluster non-convex data, difficulty in completing adaptive processing, unfavorable real-time processing of passive terahertz images, etc., to achieve self-adaptability , improve the quality of terahertz images, and have the effect of adaptability and noise robustness

Active Publication Date: 2020-09-25
NANJING UNIV OF AERONAUTICS & ASTRONAUTICS
View PDF3 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, because K-Means is sensitive to noise and outliers, as well as artificially setting the k value, it cannot accurately cluster non-convex data, so this technical method is often only used in active terahertz images, that is, image samples with better quality In the case of a certain situation, a more accurate target segmentation and detection effect can be obtained, and manual intervention is required, and it is difficult to complete adaptive processing
[0005] The traditional DBSCAN algorithm in the existing technology still has a certain degree of manual intervention, which is not conducive to the real-time processing of passive terahertz images

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
  • Real-time segmentation and classification method of passive terahertz image target
  • Real-time segmentation and classification method of passive terahertz image target
  • Real-time segmentation and classification method of passive terahertz image target

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0048] Embodiment: overall scheme flowchart of the present invention is as figure 1 As shown, the steps are:

[0049] (1) Enhance the passive terahertz image, filter out noise, and improve the quality of the terahertz image;

[0050] (2) Perform active contour processing on the terahertz image after image enhancement and optimization, detect human body and object contours, and perform contour interpolation to extract the overall contour data of the scene target;

[0051] (3) The adaptive DBSCAN algorithm is used to process the contour data to accurately output the contour segmentation results of the terahertz image target.

[0052] In step (1), Gaussian filtering, bilateral filtering, and sharpening enhancement are performed on the passive terahertz image to suppress noise to the greatest extent while retaining effective details to obtain a higher-quality terahertz image.

[0053] Such as figure 2 As shown, two pictures (a) and (b) are included, (a) is the original passive...

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 real-time segmentation and classification method for a passive terahertz image target, which belongs to the field of terahertz image processing, and has the technical key points that image enhancement is carried out on a passive terahertz image, noise is filtered, details are reserved, and a terahertz image with higher quality is obtained; geometric active contour (CV) processing is performed on the enhanced terahertz image, and edge contour data of the whole scene are obtained through contour interpolation; according to the method, adaptive density space clustering (DBSCAN) algorithm processing is carried out on contour data, and region segmentation and classification of a scene target are accurately obtained in real time under the condition that manual additionalparameters are not needed.

Description

technical field [0001] The invention belongs to the field of terahertz image processing, in particular to a real-time segmentation and classification method for passive terahertz image targets. Background technique [0002] Terahertz (THz for short) waves generally refer to electromagnetic waves with a frequency of 0.1-10 THz (wavelength of 0.03-3mm). Terahertz waves have strong penetrating ability and can perform perspective imaging on opaque objects. Compared with X-ray imaging and ultrasonic imaging, terahertz imaging has the advantages of low energy and high contrast for imaging low-density objects, and is widely used in security imaging and non-destructive testing. [0003] Passive terahertz imaging technology does not need an additional terahertz radiation source, and uses the terahertz waves emitted by the natural body itself to transmit information, and the terminal detector collects and reconstructs the information to complete the imaging. Compared with active ter...

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 Applications(China)
IPC IPC(8): G06T7/13G06T7/12G06T7/11G06T5/00G06K9/00G06K9/46G06K9/62
CPCG06T7/13G06T7/12G06T7/11G06T2207/20081G06V40/10G06V10/44G06F18/23G06T5/70
Inventor 徐帆蒲子琪刘娅
Owner NANJING UNIV OF AERONAUTICS & ASTRONAUTICS
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
Eureka Blog
Learn More
PatSnap group products