Search

CN-115575373-B - Correction method and correction device for spectroscopic laser-induced phosphorescence imaging system

CN115575373BCN 115575373 BCN115575373 BCN 115575373BCN-115575373-B

Abstract

The invention discloses a correction method and a correction device for a spectroscopic laser-induced phosphorescence imaging system. The calibration method comprises the steps of selecting a light homogenizing plate as a standard light source, enabling the light emitting surface of the light homogenizing plate to be located at a focusing surface of an imaging system, setting acquisition frequency, acquisition total time and N different exposure times, acquiring N groups of gray images acquired by the imaging system under different exposure times, averaging the acquired N groups of gray images according to pixels to obtain an image gray matrix of each group of gray images, performing linear fitting on a gray matrix set under the N exposure times by using a least square method to obtain a bias correction matrix and a gain correction matrix, and performing correction processing on an original response gray matrix acquired under a test state according to the obtained bias correction matrix and the gain correction matrix.

Inventors

  • XU CHUANLONG
  • Chen Manfu
  • ZHANG BIAO
  • LI JIAN

Assignees

  • 东南大学

Dates

Publication Date
20260505
Application Date
20221031

Claims (7)

  1. 1. A method for calibrating a spectroscopic laser-induced phosphorescence imaging system, comprising: the method comprises the steps of selecting a light homogenizing plate with the size full of the whole view field of a CCD camera as a standard light source, placing an adjusted beam-splitting laser-induced phosphorescence imaging system in front of the light homogenizing plate, and enabling the light emitting surface of the light homogenizing plate to be positioned at a focusing surface of the imaging system; Setting acquisition frequency, acquisition total time and N different exposure times, and acquiring N groups of gray images acquired by an imaging system under different exposure times; Averaging the acquired N groups of gray images according to pixels to obtain an image gray matrix set (Y 1 Y 2 ...Y N ) of each group of gray images; Performing linear fitting on a gray matrix set (Y 1 Y 2 ...Y N ) under N exposure times by using a least square method to obtain a bias correction matrix and a gain correction matrix; and correcting the original response gray matrix obtained in the test state according to the obtained bias correction matrix and gain correction matrix.
  2. 2. The correction method of spectroscopic laser-induced phosphorescence imaging system of claim 1, wherein the step of obtaining the bias correction matrix and the gain correction matrix by linear fitting using a least square method on the gray matrix set (Y 1 Y 2 ...Y N ) at N exposure times, comprises: Performing linear fitting on the pixel point Y max with the maximum gray value of the image gray matrix set (Y 1 Y 2 ...Y N ) by using a least square method; determining bias errors and gain coefficients of the ideal response curve according to the fitting result; determining bias errors, gain coefficients and correction factors of gain coefficients of response curves of any pixel Y i,j in an image gray matrix Y o according to bias errors and gain coefficients of ideal response curves, wherein i is an integer and is 1-1 m, j is an integer and is 1-1 j N, and o is an integer and is 1-1 o N; And obtaining a bias correction matrix and a gain correction matrix according to the bias error, the gain coefficient and the correction factor of the gain coefficient of the response curve of any pixel Y i,j in the obtained image gray matrix Y o .
  3. 3. The method for calibrating a spectroscopic, laser-induced phosphorescence imaging system of claim 2, The method comprises the steps of performing linear fitting on a pixel point Y max with the maximum gray value of an image gray matrix (Y 1 Y 2 ...Y N ) by using a least square method, wherein the linear fitting formula is as follows: Where a max is the response gain factor, b max is the response bias error, r max is the linear fitness, x is the constant light source intensity value, and the remaining variables are: Where x N represents the exposure time of the nth set of gray scale images and y maxN represents the maximum value of the gray scale of the pixels in the nth set of gray scale matrices.
  4. 4. The method of calibrating a spectroscopic, laser-induced phosphorescence imaging system of claim 3, wherein the step of determining the bias error and gain factor for the ideal response curve based on the fitting result comprises: taking y maxN to represent the maximum value of pixel gray in the N group gray matrix and taking the maximum value as a calibration point; the bias error and gain coefficient of the ideal response curve are determined as follows: b max ＝(y max1 +y max2 +...+y maxN )/N-a max (x 1 +x 2 +...+x N )/N a max ＝((y max1 ·x 1 +y max2 ·x 2 +...+y maxN ·x N )/N-(y max1 +y max2 +...+y maxN )·(x 1 +x 2 +...+x N )/N 2 )/((x 1 2 +x 2 2 +...+x N 2 )/N-(x 1 +x 2 +...+x N ) 2 /N 2 ) Wherein b max is the bias error of the ideal response curve, a max is the gain coefficient of the ideal response curve; determining bias errors and gain coefficients of response curves of any pixel in the image gray matrix Y o , wherein the bias errors and the gain coefficients are respectively as follows: b i,j ＝(y i,j,1 +y i,j,2 +...+y i,j,N )/N-a i,j (x 1 +x 2 +...+x N ) a i,j ＝((y i,j,1 ·x 1 +y i,j,2 ·x 2 +...+y i,j,N ·x N )/N-(y i,j,1 +y i,j,2 +...+y i,j,N )·(x 1 +x 2 +...+x N )/N 2 )/((x 1 2 +x 2 2 +...+x N 2 )/N-(x 1 +x 2 +...+x N ) 2 /N 2 ) Wherein, b i,j is the offset error of the response curve, a i,j is the gain coefficient of the response curve, and y i,j,N is the pixel in the Nth group of gray level images; Determining a correction factor c i,j of a gain coefficient of a response curve of any pixel in the image gray matrix Y o : c i,j ＝a max /a i,j and (3) obtaining a bias correction matrix B and a gain correction matrix C for the gray matrix set (Y 1 Y 2 ...Y N ) under N exposure times, wherein B is a set of B i,j , and C is a set of C i,j .
  5. 5. The method of any one of claims 1-4, wherein any one of the set of image gray matrices Y o (Y 1 Y 2 ...Y N ) is: wherein o is an integer and 1≤o≤N.
  6. 6. The method for correcting the response of a spectroscopic laser-induced phosphorescence imaging system in a large scale according to claim 5, wherein in the step of correcting the gray matrix of the original response obtained in the test state according to the obtained bias correction matrix and gain correction matrix, the correction formula is: Y correction =(Y-B).*C Wherein Y correction is the corrected gray matrix, B is the offset correction matrix, C is the gain correction matrix, and Y is the original response gray matrix obtained in the test state.
  7. 7. A spectroscopic laser induced phosphorescence imaging system correction apparatus comprising a processor and a memory, the memory having stored therein a program or instructions that are loaded and executed by the processor to implement the steps of the spectroscopic laser induced phosphorescence imaging system correction method of any one of claims 1 to 6.

Description

Correction method and correction device for spectroscopic laser-induced phosphorescence imaging system Technical Field The invention relates to the technical field of measurement, in particular to a response large-range correction method of a spectroscopic laser-induced phosphorescence imaging system. Background The high-temperature gas temperature before the turbine of the aero-engine has important influence on thrust-weight ratio, engine efficiency and unit oil consumption emission, and the high-temperature gas temperature field is required to be accurately measured, but the application of the traditional temperature measurement technology is severely limited by the complicated flow field. The laser-induced phosphorescence (LIP) temperature measurement technology has the characteristics of non-invasiveness, less required optical window, high reliability and the like, and has great application potential in the aspects of optical window limitation, temperature field measurement involving complex flow and heat exchange processes. The Laser Induced Phosphorescence (LIP) temperature measurement technique is a non-contact optical temperature measurement technique based on phosphorescence thermal quenching effect. The current mature laser-induced phosphorescence temperature measurement method comprises an absolute intensity method, a life attenuation method and an intensity ratio method. The absolute intensity method is used for measuring the temperature based on the characteristic that the phosphorescence intensity is reduced along with the increase of the ambient temperature, is suitable for a temperature measurement scene of uniformly distributing phosphorescence substances, such as spraying a phosphorescence coating with uniform thickness on a high-temperature solid wall surface, and indirectly measures the temperature distribution of the high-temperature solid wall surface by measuring the phosphorescence intensity of the phosphorescence coating. The life attenuation method realizes temperature measurement based on the characteristic that the phosphorescence life is reduced along with the increase of the ambient temperature, a high-speed camera is generally used for continuously collecting phosphorescence signals of the same detected area, phosphorescence life distribution is obtained by fitting, and the temperature distribution of a solid wall surface is obtained by inversion according to the relationship between phosphorescence life and the ambient temperature. The intensity ratio rule is to measure the temperature by utilizing the relation between the ratio of the phosphorescence intensities of two different wave bands in the phosphorescence emission spectrum and the ambient temperature, and can be used for temperature measurement scenes with complex flow or uneven distribution of phosphorescence particles. The intensity ratio method in the laser-induced phosphorescence temperature measurement method is most suitable for measuring a high-temperature gas temperature field. Based on this technology, a spectroscopic laser induced phosphorescence imaging system has been developed in recent years, as shown in fig. 1. The system consists of a dichroic mirror, a cage plate (fixed dichroic mirror), a filter plate, a cage plate (fixed filter plate), a lens, a CCD camera and the like. In order to verify the feasibility of measuring the gas temperature field by a laser-induced phosphorescence intensity ratio method, a laboratory designs and builds a high-temperature jet experiment table with T less than or equal to 500 ℃, and BAM: EU 2+ phosphorescence particles are selected as a thermal imaging medium for temperature field measurement. The emission spectrum of EU 2+ after excitation by 355nm laser is shown in FIG. 2. As can be seen from the graph, the phosphorescence wavelength is mainly 400-500nm, wherein the phosphorescence intensity of 425+/-15 nm increases with the temperature, and the phosphorescence intensity of 466+/-18 nm is basically unchanged with the temperature. The intensity value of the phosphorescence of 425+/-15 nm is represented by I 425, the intensity value of the phosphorescence of 466+/-18 nm is represented by I 466, and the temperature measurement by the laser-induced phosphorescence intensity ratio method is based on the relation between the phosphorescence intensity ratio I 425/I466 and the temperature, as shown in the formula (1). The phosphorescence of 400-500nm emitted by the phosphorescence particles is divided into two wave bands of 400-445nm and 445nm-500nm by the dichroic mirror, wherein the phosphorescence of 400-445nm is imaged on the CCD camera 1 after being reflected by 45 DEG mirror surface of the dichroic mirror and filtered by a filter of 425+/-15 nm, and the phosphorescence of 445nm-500nm is imaged on the CCD camera 2 after being filtered by a filter of 466+/-18 nm by the dichroic mirror. The temperature field can be inverted by processing the gray values of two different-band