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

A Non-invasive Latent Fingerprint Extraction Method Based on Hyperspectral Image

A hyperspectral image, non-invasive technology, applied in the field of non-invasive latent fingerprint extraction, which can solve the problems of complex background information, destruction of the target to be detected, invisible latent fingerprints, etc.

Active Publication Date: 2021-10-29
RES & DEV INST OF NORTHWESTERN POLYTECHNICAL UNIV IN SHENZHEN
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] The purpose of the present invention is to provide a non-invasive latent fingerprint extraction method based on hyperspectral images, which solves the problem of invisible fingerprints and generally complex background information, and at the same time, it is not allowed to destroy the target to be detected in many occasions. Potential Fingerprinting Problems Under Difficulty

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
  • A Non-invasive Latent Fingerprint Extraction Method Based on Hyperspectral Image
  • A Non-invasive Latent Fingerprint Extraction Method Based on Hyperspectral Image
  • A Non-invasive Latent Fingerprint Extraction Method Based on Hyperspectral Image

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0055] Such as figure 1 As shown, a non-invasive latent fingerprint extraction method based on hyperspectral images is characterized in that: comprising the following steps carried out in sequence:

[0056] S1. Collect hyperspectral data of the object to be detected to obtain multi-channel data Among them, h, w, and L represent the height, width, and number of channels of the hyperspectral image data, respectively, Represents a three-dimensional real number space, and preprocesses the collected data to obtain

[0057] S2, for preprocessing data Perform dimensionality reduction processing to get

[0058] S3, for the kth (k=1,...,l) channel image X k Calculate the local total variation and separate the non-textured part U from the image k and texture part V k ;

[0059] S4. Fusing the multi-channel texture images obtained in S3 to obtain a fingerprint image extracted after fusing the l-channel images.

Embodiment 2

[0061] The difference between this embodiment and Embodiment 1 is that further, the method of preprocessing the collected data in step S1 includes at least one of eliminating data noise through low-pass filtering and logarithmic transformation to enhance dark area details .

[0062] Further, the dimensionality reduction processing method in step S2 includes at least one of principal component analysis, linear discriminant analysis, and direct extraction of partial channels.

[0063] Further, in the step S3, for the kth (k=1,...,l) channel image X k Calculate the local total variation and separate the non-textured part U from the image k and texture part V k The method includes the following steps:

[0064] S301. Calculate image X k The local total variation of , and denoted as LTV σ (X (k) );

[0065]

[0066] where L σ is a low-pass filter with cutoff radius σ, is the gradient operator;

[0067] S302. Calculate image X k The texture intensity λ of the (i, j) pi...

Embodiment 3

[0092] Concrete flow process of the present invention is as follows:

[0093] S1, data collection and preprocessing, the specific steps are as follows:

[0094] S101. Collect 31 channels of data with a size of 512*640 through hyperspectral imaging technology;

[0095] S102. Due to the presence of noise in the data acquisition, the noise is removed by Gaussian low-pass filtering;

[0096] S103 enhances the contrast of dark area details through logarithmic transformation, and sharpens the fingerprint image;

[0097] S2. Dimensionality reduction processing, the specific steps are as follows:

[0098] S201. In the present invention, principal component analysis is used for data dimensionality reduction, and the reconstruction threshold is set to 95%, to obtain the first four principal component channels;

[0099] S3, single-channel texture information extraction, the specific steps are as follows: the present invention uses the calculation of local total changes to separate the...

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 non-invasive latent fingerprint extraction method based on a hyperspectral image, which involves firstly using a hyperspectral imager to collect data in an area to be detected, obtaining multi-channel data, and performing data pre-processing through low-pass filtering and logarithmic transformation. processing; then reduce the dimensionality of the multi-channel data to obtain low-dimensional data; finally calculate the local total variation of each channel image to separate the texture and non-texture information, and fuse the texture information of each channel through the gradient statistical histogram information to obtain fingerprint image. Its advantages are: the present invention utilizes rich spectral information and texture information, overcomes the problem of poor detection effect of traditional fingerprint extraction methods under complex background images, and has better detection effect under the condition of completely protecting the object to be detected. It has broad application prospects in criminal investigation, cultural relic detection, information security and other fields.

Description

technical field [0001] The invention relates to the field of information security, in particular to a non-invasive latent fingerprint extraction method based on a hyperspectral image. Background technique [0002] Among the many biometrics, fingerprint recognition technology has become the most widely used technology due to its high practicability and reliability, and it has legal effect. Fingerprint feature recognition has a long history of development from its earliest application in the field of judicial criminal investigation to its wide application in the civilian field. Fingerprint identification, in a nutshell, relies on the unique feature formed by the fingerprint texture that can represent a person's identity, and associates it with an individual. By comparing the personal fingerprint feature with the pre-saved library fingerprint feature, the authentication and identification of personal identity is realized. . As the most traditional and mature identification me...

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): G06K9/00
CPCG06V40/1312G06V40/1347G06V40/1365
Inventor 陈捷阎龙斌
Owner RES & DEV INST OF NORTHWESTERN POLYTECHNICAL UNIV IN SHENZHEN
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