Method and apparatus for single-pixel thermal imaging detection of surface and internal defects of material

Patent metadata

Field	Value
Publication number	US-12513376-B2
Application number	US-202418829827-A
Country	US
Kind code	B2
Filing date	Sep 10, 2024
Priority date	May 31, 2024
Publication date	Dec 30, 2025
Grant date	Dec 30, 2025

Abstract

Provided are a method and apparatus for single-pixel thermal imaging detection of surface and internal defects of a material. The laser module is configured to generate uniform laser light. The spatial light modulation module is configured to spatially encode and modulate received uniform laser light. The projection lens module is configured to project and amplify an encoded light field, and project the encoded light field onto a surface of an object under detection. The optical convergence coupling module is configured to capture and image thermal radiation of the object under detection, and integrate and sum thermal radiation intensities of a projection heating area to obtain a thermal radiation temperature. The thermal infrared single-pixel detector is configured to measure the thermal radiation temperature. The image reconstruction module is configured to sparsely reconstruct data, and finally reconstruct a defect detection result of the projection heating area.

First claim

What is claimed is:

1 . A method for single-pixel thermal imaging detection of surface and internal defects of a material, comprising the following steps:
S1: generating laser light in a near-infrared band; S2: generating m mutually independent Bernoulli random matrices with a size of √{square root over (n)}×√{square root over (n)}, wherein n represents a number of elements of one of the Bernoulli random matrices, and m represents a number of the Bernoulli random matrices; S3: measuring a temperature field of a to-be-projected heating area on a surface of an object under detection, and summing measured temperatures to obtain an initial temperature value T i0 , wherein i=1, 2, . . . , m; S4: spatially encoding the laser light by using one of the Bernoulli random matrices generated in step S2, projecting the spatially encoded laser light to a projection heating area on the surface of the object under detection, and heating the surface of the object under detection by using the spatially encoded laser light; S5: measuring a temperature field of the projected heating area on the surface of the object under detection, summing measured temperatures to obtain a heating temperature value T i , and calculating T di =T i −T i0 , wherein i=1, 2, . . . , m; S6: removing heating laser light on the surface of the object under detection, that is, skipping heating the object under detection, and keeping for a time t h , so as to restore the object under detection to be close to the initial temperature before heating; S7: repeating steps S3 to S6, sequentially using different ones of the Bernoulli random matrices, and forming vectors T d of m×1 by using T di calculated each time; S8: sequentially vectorizing the m Bernoulli random matrices to sequentially form a row vector of 1×n, a row vector of 1×n formed by vectorization of the i th Bernoulli random matrix being expressed as ϕ i , and then forming a matrix Φ by all row vectors, wherein Φ∈R m×n , R m×n represents a real number set matrix of m×n, and an i th row of the matrix Φ is a row vector ϕ i ; S9: generating an inverse discrete cosine matrix Ψ of n×n, setting a to-be-solved sparse vector x, wherein x∈R n×1 , and R n×1 represents a real number set vector of n×1, and forming a target problem ΦΨx=T d , S10: solving the following objective function by using an optimization algorithm:

min
⁢


x


1

⁢

s
.
t
.

ΦΨ

⁢
x

=

T
d

wherein ∥⋅∥ 1 represents a vector 1-norm, that is, the sum of absolute values of all elements; and
S11: after the sparse vector x is obtained, calculating a to-be-reconstructed data vector b=Ψx, and matrixing the obtained vector b to obtain a reconstructed detection image.

2 . The method for single-pixel thermal imaging detection of surface and internal defects of a material according to claim 1 , wherein in step S4, before the laser light is spatially encoded, the laser light is sequentially homogenized, collimated and amplified.

3 . The method for single-pixel thermal imaging detection of surface and internal defects of a material according to claim 1 , wherein in step S5, being close to the initial temperature before heating means being within 1.5° C. from the initial temperature before last heating.

4 . An apparatus for single-pixel thermal imaging detection of surface and internal defects of a material applied to the method according to claim 1 , comprising a laser module, a spatial light modulation module, a projection lens module, an optical convergence coupling module, a thermal infrared single-pixel detector and an image reconstruction module, wherein
the laser module is configured to generate laser light with a uniform radiation density and provide an initial uniform light source for the spatial light modulation module; the spatial light modulation module is configured to spatially encode and modulate received uniform laser light and project modulated laser light to the projection lens module; the projection lens module is configured to project and amplify an encoded light field modulated by the spatial light modulation module, and project the encoded light field onto a surface of an object under detection, which is equivalent to heat flow input and causes heat conduction inside the material; the optical convergence coupling module is configured to capture and image thermal radiation of the object under detection caused by a temperature change, and integrate and sum thermal radiation intensities of a projection heating area to obtain a thermal radiation temperature; the thermal infrared single-pixel detector is configured to measure the thermal radiation temperature to obtain a thermal radiation temperature value; and the image reconstruction module is configured to sparsely reconstruct data measured by the thermal infrared single-pixel detector, and finally reconstruct a defect detection result of the projection heating area.

5 . The apparatus for single-pixel thermal imaging detection of surface and internal defects of a material according to claim 4 , wherein the laser module comprises a laser generator, an end collimator and a laser beam expander, the laser generator is configured to generate near-infrared band laser light, the laser light is transmitted to the end collimator through optical fiber coupling, the laser light subjected to optical fiber coupling has a uniform emergent radiation density, the end collimator is configured to collimate the uniform laser light, and the laser beam expander is configured to amplify parallel light with a small diameter collimated by the end collimator.

6 . The apparatus for single-pixel thermal imaging detection of surface and internal defects of a material according to claim 4 , wherein the spatial light modulation module comprises a spatial light modulator and a driving board, the spatial light modulator is configured to spatially encode and modulate the received uniform laser light and is controlled by the driving board; and the driving board is responsible for providing a power supply and control signals required by the spatial light modulator, and after being controlled, the spatial light modulator spatially encodes and modulates the uniform laser light and projects modulated laser light to the projection lens module.

7 . The apparatus for single-pixel thermal imaging detection of surface and internal defects of a material according to claim 4 , wherein a Bernoulli random matrix is adopted for the spatial encoding and modulation.

8 . The apparatus for single-pixel thermal imaging detection of surface and internal defects of a material according to claim 4 , wherein the projection lens module is a thermal infrared imaging lens.

9 . The apparatus for single-pixel thermal imaging detection of surface and internal defects of a material according to claim 4 , wherein the optical convergence coupling module comprises an integrating lens, the integrating lens is configured to integrate and sum the thermal radiation intensities of the projection heating area to obtain the thermal radiation temperature, and a cold stop capable of shielding an area outside the project

Assignees

• Univ Hunan

Inventors

• Wang Hongjin
• He Yunze
• Li Xiang
• Deng Baoyuan
• Wang Yaonan

Classifications

• G06T2207/10048

  Infrared image · CPC title

• G06T7/0004

  Industrial image inspection · CPC title

• G01J3/2823

  Imaging spectrometer · CPC title

• H04N23/21 Primary

  from near infrared [NIR] radiation · CPC title

• G01J2005/0077

  Imaging · CPC title