CN111879730A - Optical coherence tomography signal processing method based on rectangular window function optimization - Google Patents
Optical coherence tomography signal processing method based on rectangular window function optimization Download PDFInfo
- Publication number
- CN111879730A CN111879730A CN202010817011.4A CN202010817011A CN111879730A CN 111879730 A CN111879730 A CN 111879730A CN 202010817011 A CN202010817011 A CN 202010817011A CN 111879730 A CN111879730 A CN 111879730A
- Authority
- CN
- China
- Prior art keywords
- signal
- phase
- window
- depth
- rectangular window
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/47—Scattering, i.e. diffuse reflection
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/41—Refractivity; Phase-affecting properties, e.g. optical path length
- G01N21/45—Refractivity; Phase-affecting properties, e.g. optical path length using interferometric methods; using Schlieren methods
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/47—Scattering, i.e. diffuse reflection
- G01N2021/4704—Angular selective
- G01N2021/4709—Backscatter
Landscapes
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
When an optical coherence tomography system is used for imaging a measured sample, the dispersion mismatch between a sample arm and a reference arm affects the system resolution, so that depth-resolved dispersion compensation is very important. Interference signals at different depths of the sample are intercepted from the A-line signal and then subjected to Fourier transform (FFT), and a rectangular window function used for intercepting is optimized, so that phase errors generated by windowing in the signal windowing FFT can be effectively reduced, dispersion coefficients of the sample to be measured at different depths are accurately extracted, and dispersion compensation precision is improved.
Description
Technical Field
The invention relates to a frequency Domain Optical coherence tomography (FD-OCT) technology, in particular to an Optical coherence tomography signal processing method based on rectangular window function optimization.
Background
Optical Coherence Tomography (OCT) is a high-resolution, non-invasive optical tomography technique that can detect microstructures inside a body by detecting the intensity of backscattered light from the sample to obtain structural information of the sample. The frequency-domain OCT technique obtains a chromatogram of a measured object through Inverse Fourier Transform (IFT) of a frequency-domain interference spectrum signal, having a spatial resolution of the order of micrometers or submicrometers. However, in high resolution systems, the dispersion effect is exacerbated by the use of ultra-broad spectrum light sources with bandwidths in excess of 100nm, and even up to several hundred nm. The dispersion causes the broadening and distortion of the OCT coherent signal, and the actual resolution of the system is smaller than the theoretical value, so the dispersion compensation is one of the key technologies for realizing the high-resolution OCT.
The dispersion algorithm compensation is to eliminate dispersion broadening by carrying out post-processing on data acquired by OCT, and has the advantages of flexibility, convenience and the like. The main methods comprise a deconvolution algorithm, an iterative algorithm, a self-focusing algorithm, a full-depth dispersion compensation method and the like. Because most of the tested tissues have dispersion with depth change, Zhang Xian Ling, Huang Paije, A.F. Fercher, etc. propose different depth-resolved dispersion compensation methods for frequency-domain optical coherence tomography, and through dynamic filtering, separate the interference spectrum of each depth of the tested sample, and further extract the dispersion coefficient of each layer depth to perform respective compensation, so as to obtain better dispersion compensation effect in the full depth range. In the method, signals at different depths of a sample need to be extracted to perform FFT analysis on phase information to obtain second-order and above dispersion coefficients at different depths, so that a window function is generally selected to intercept a sample structure signal to obtain a signal at a certain depth, and then subsequent signal processing is performed. When depth resolution dispersion compensation is carried out, interference signals at different depths are required to be separated, only the main lobe frequency is required to be accurately read after windowing, and amplitude precision is not considered; the rectangular window is narrow in main lobe, large in side lobe, highest in frequency identification accuracy and lowest in amplitude identification accuracy, so that the rectangular window is used. However, in the process of windowing and intercepting signals, if the window width is too small, information of some structural layers may be lost, and if the window width is too wide, the effect of the post-processing depth dispersion compensation algorithm may be weakened. In addition, in order to analyze the spectrum information at different depths more accurately, when the interference signal at the local depth is intercepted by windowing, the window width needs to be increased as much as possible in principle, so as to improve the frequency resolution of the spectrum. Suitable window selection is therefore very important in the phase extraction of the interference signal at different depths.
Disclosure of Invention
The invention aims to provide an optical coherence tomography signal processing method based on rectangular window function optimization. When an optical coherence tomography system is used for imaging a measured sample, the dispersion mismatch between a sample arm and a reference arm affects the system resolution, so that depth-resolved dispersion compensation is very important. At the moment, interference signals at different depths of a sample are intercepted from an A-line signal and then subjected to FFT, a rectangular window function used for intercepting is optimized, namely, the interference signals obtained by a primary rectangular window are subjected to FFT to extract phases, the obtained phases are fitted with a quadratic polynomial by a least square method according to the functional relation between the phases and wave numbers when dispersion exists, the length and the central position of the rectangular window are continuously changed by taking the standard error of the fitted quadratic polynomial as a judgment condition, wherein the minimum standard error corresponds to the optimized rectangular window, and the approximate real phase can be obtained after the interference signals intercepted under the rectangular window are subjected to FFT. Phase errors caused by introduction of the rectangular window are reduced, and dispersion coefficients of the measured sample at different depths are extracted more accurately, so that dispersion compensation precision is improved.
The technical solution of the invention is as follows:
an optical coherence tomography signal processing method based on rectangular window function optimization is characterized by comprising the following steps:
firstly, a frequency domain optical coherence tomography system is utilized to scan a sample, and a photoelectric detection array of the system records an original interference signal of the sample and transmits the signal to a computer;
secondly, performing inverse Fourier transform on the original interference signal after background removal to obtain an A-line signal;
thirdly, adding a primary rectangular window to a signal in any depth in the A-line signal to intercept the signal, and after Fourier transform is carried out on the intercepted signal, calculating a phase angle to obtain a primary phase corresponding to the depth signal;
fourthly, according to the functional relation between the phase and the wave number during dispersion, the primary phase is taken as a target function, quadratic polynomial of the primary phase and the wave number is obtained by fitting with a least square method, and the standard error of the fitting result is calculated;
changing the window length and the central position of the primary rectangular window to obtain the current rectangular window; intercepting a signal in the depth range by adding a current rectangular window to the A-line signal, performing FFT (fast Fourier transform) on the intercepted signal, then solving a phase angle to obtain a phase corresponding to the depth signal, taking the obtained phase as a target function, fitting by using a least square method to obtain a quadratic polynomial of the phase and a wave number, and calculating a standard error of a fitting result, namely the standard error of the phase corresponding to the rectangular window; then, continuously changing the window length and the central position of the rectangular window, intercepting signals in the depth range, and obtaining standard errors of phases corresponding to different rectangular windows after the rectangular windows traverse all set conditions;
sixthly, the rectangular window corresponding to the minimum standard error value obtained in the fifth step is the optimal rectangular window, and the optimal rectangular window is utilized to obtain the final phase corresponding to the depth signal;
seventhly, calculating a corresponding second-order dispersion coefficient according to the final phase of the depth signal;
sixthly, repeating the step of the signals with different depths in the A-line signal to obtain second-order dispersion coefficients corresponding to the signals with different depths, and calculating phase deviation caused by dispersion; subtracting the deviation value from the phase corresponding to the depth to obtain the phase after dispersion compensation;
and ninthly, reconstructing the frequency domain interference signal after dispersion compensation by utilizing the phase after each depth compensation.
The method for selecting the window length and the central position of the primary rectangular window in the third step comprises the following steps:
when no sample exists, the average amplitude of the A-line signal is the mean value of noise;
when a sample is set, the maximum value of a signal with a certain depth in the A-line signal takes the depth of the maximum value as the central position of the primary rectangular window, and the window is expanded towards two sides until the signal intensity of the two sides is equal to the mean value of noise, and at the moment, the window length of the primary rectangular window is the window length of the depth signal.
The method for changing the window length and the center position of the rectangular window in the fifth step comprises the following steps:
taking the unit length of the A-line signal as a step length, moving the initial position of the primary rectangular window to the right by 1 step length, sequentially moving the tail end position of the window to the left and the right from 1 step length to N step lengths to obtain 2N rectangular windows in total, intercepting an interference signal under each window to perform Fourier transform to extract a phase, and repeating the operation until the initial position of the window moves to the right by N step lengths; similarly, the initial position of the primary rectangular window is moved by 1 step left, the end position of the window is sequentially moved from 1 step left to N step right to obtain 2N rectangular windows, interference signals under each window are intercepted, Fourier transform is carried out to extract the phase, the process of moving the window initial position by N steps left is repeated until the end, the phase corresponding to the depth signal is obtained when all different rectangular windows intercept signals, and N is 15-30.
In the step (c), a second-order dispersion coefficient corresponding to the depth is calculated, specifically:
step 7.1 Fourier transform is carried out on the interference signal at the depth to obtain final phase information
Wherein k represents a wave number,. DELTA.znIs the optical path difference, k, of the sample n-th layer with respect to the reference arm mirror0Wave number corresponding to central wavelength of light source，nnIs the effective refractive index at the nth layer of the sample, ng,nIs the effective group index at the nth layer of the sample, a2Is the second order dispersion compensation coefficient at the nth layer of the sample.
Step 7.2 with (k-k)0) As independent variable, pair
And performing multiple term numerical fitting to obtain a quadratic phase term, namely a second-order dispersion coefficient corresponding to the signal at the depth.
The optical coherence tomography signal processing method based on rectangular window function optimization is characterized in that when depth resolution dispersion compensation is carried out on a measured sample, windowed FFT is carried out on interference signals at different depths of the sample, the window length and the central position of a primary rectangular window function determined by a single threshold value used for A-line signal interception are optimized, the true value of a phase is continuously approached, and phase errors introduced by windows in the windowed FFT are reduced, so that the optimal dispersion coefficients of the sample at different imaging depths are obtained, and a better dispersion compensation effect is achieved.
The technical solution principle of the invention is as follows:
at present, a method for extracting time-frequency domain phase of a digital interference signal is mainly a Fourier transform method, but the Fourier transform is a global transform and is more suitable for analyzing a steady signal; when analyzing non-stationary signals, spectral aliasing of different local regions is generated, so that the fundamental frequency cannot be accurately extracted. The windowed Fourier transform can effectively localize the frequency spectrum, improve the superposition phenomenon of different levels of frequency spectrums, and improve the measurement accuracy, but if the window size in the windowed Fourier transform is fixed, errors can be introduced in the analysis of frequency spectrum information.
An ideal FFT requires that the time domain signal is infinitely long, while in practical computational measurements, only a finite length signal can be FFT-ed, which is equivalent to truncating the infinitely long signal. If the input signal is infinitely long, the resulting spectrum from the FFT is the exact correct spectrum, but if the input signal is a truncated finite-length recorded sample and the truncated signal is not a full period, then the spread of the discrete spectral lines on both sides produces spectral leakage. Because the Fourier transform spectrum of the rectangular function is infinitely expanded in two directions, a ringing phenomenon can be generated, and a side lobe effect can be generated by calculating the spectrum by using an FFT algorithm after a rectangular window is added to a signal. Therefore, spectral line blurring and spectral resolution reduction are caused by spectral leakage generated by windowing interception, and the leakage degree is reduced along with the increase of the interception length, and the longer the interception length is, the smaller the leakage is, and the higher the spectral resolution is. When the interception length is short, because the rectangular window function spectrum has a plurality of side lobes, a plurality of side lobes are formed after the window function spectrum is convoluted with the main signal spectrum, and the side lobes play a role of interference between spectrums and also reduce the spectrum resolution.
When the depth-resolved dispersion compensation is carried out on the optical coherence tomography of the sample, the window length and the central position of a proper rectangular window are required to be selected to intercept interference signals at different depths. If the window width is too small, information of some structural layers can be lost, if the window width is too wide, the window width can be too wide, the depth dispersion compensation effect is weakened, and because the truncated length is longer for reducing the spectrum leakage to improve the spectrum resolution, the selection of the proper rectangular window length and the center position is crucial, so that a more real spectrum can be restored.
The imaging is carried out by detecting the back scattering light of a sample in the frequency domain OCT, the low coherent light emitted by a light source is respectively irradiated on a reflector and the sample through a reference arm and a sample arm, the reference light returned from the reflector interferes with the sample light returned from different depths of the sample, a direct current term is removed, and the interference signal of the sample is as follows:
where k denotes the wave number, Re denotes the real part of the complex number, In(k) Representing the intensity of light scattered back from the nth layer of the sample, Ir(k) Representing the intensity of light returned by the mirror, Δ znIs the optical path difference of the sample nth layer relative to the reference arm mirror,
is the phase difference of the scattered light of the nth layer of the sample relative to the reference light, including the higher-order dispersion phase phi (k, deltaz)n). The introduction of high-order dispersion phase is the main reason of leading to the envelope broadening and distortion of interference signals and reducing the resolution of a system, and the purpose of dispersion compensation is to eliminate the high-order dispersion phase.
Phase difference of scattered light of nth layer relative to reference light
Can be expressed as:
in the formula betan(k) Is the effective propagation coefficient at the nth layer of the sample, beta for an OCT system for which the optical elements in the reference and sample arms have been givenn(k) Wave number k corresponding to the wavelength at the center of the light source0And performing Taylor series expansion on the vicinity, thus obtaining:
wherein n isnIs the effective refractive index at the nth layer of the sample, ng,nIs the effective bulk refractive index, beta, at the nth layer of the sample "n、β”'nRespectively, the second order effective dispersion coefficient and the third order effective dispersion coefficient at the nth layer of the sample, a2、a3Respectively called second-order dispersion compensation coefficient and third-order dispersion compensation coefficient.
The samples have different dispersion coefficients beta at different depths "nAnd beta'n. To realize depth-resolved dispersion compensation, z needs to be obtained firstnThe phase position is obtained by IFT transforming the frequency domain OCT interference spectrum signal to obtain the A-line signal of the sample, and then intercepting the A-line signal through a rectangular window function to obtain z signals with different depthsnEnvelope of the structural layer signal, FFT of the envelope signal along the depth z-axis to obtain different depths znThe interference spectrum signal of (2) is extractedTo the phase. Will znAnd performing numerical fitting on the phase information to obtain second-order and third-order effective dispersion coefficients, calculating the phase introduced by dispersion, and subtracting the phase from the total phase to obtain a compensated interference spectrum signal. And then carrying out inverse Fourier transform on the compensated interference spectrum signal to obtain a sample non-dispersive A-line signal.
Generally, in an actual optical coherence tomography system, the influence of third-order dispersion and higher-order dispersion is small, and is negligible compared with the influence of a second-order dispersion term, so that dispersion compensation mainly eliminates second-order dispersion phase existing in an interference signal, and a phase difference obtained by the above formula can be expressed as:
it can be seen that the theoretical phase is a second order polynomial in wavenumber. In the traditional method, a window determined by a single threshold is used for intercepting an A-line signal, and in this case, the signal intercepted at some depth has errors in the extracted phase due to improper window length and central position, so that the calculation accuracy of a subsequent algorithm is influenced. Therefore, in order to more accurately extract the phase of a sample at different depths, a primary rectangular window is used for intercepting an A-line signal to obtain a certain depth phase, the obtained phase is fitted with a quadratic polynomial by a least square method according to the functional relation between the phase and wave number when dispersion exists, the length and the central position of the rectangular window are continuously changed by taking the standard error of the fitted quadratic polynomial as a judgment condition, when the standard error is minimum, the optimal rectangular window is corresponding, and the rectangular window is used for intercepting an interference signal and then carrying out FFT to obtain an approximate real phase. The primary rectangular window is determined by finding the maximum value of the A-line signal at the depth higher than the maximum value of the noise, and expanding the window towards two sides by taking the maximum value as the center until the signal intensity at two sides is equal to the mean value of the noise, wherein the window width at the moment is the primary window width at the maximum value, and the center position is the maximum value.
The invention has the following advantages:
in the depth dispersion compensation of the optical coherence tomography, more accurate phase information is obtained by optimizing a rectangular window function, and the phase error introduced by the windowing of interference signals can be effectively reduced.
The method is also suitable for optimizing other window functions in the phase analysis and extraction of the optical coherence tomography interference signal.
Drawings
FIG. 1 is an A-line plot of the structure of a simulated sample of the present invention, wherein (a) is a non-dispersive A-line plot and (b) is a dispersive A-line plot.
Fig. 2 is a phase error map of different depths when a simulated sample is free of dispersion, wherein (a) the phase error map is obtained after signals are intercepted and processed for a primary rectangular window, and (b) the phase error map is obtained after signals are intercepted and processed for an optimized window.
FIG. 3 is a phase error plot for different depths obtained by intercepting and processing signals with a primary rectangular window when a simulated sample has chromatic dispersion.
Fig. 4 is a phase diagram of a simulated sample with dispersion as a function of depth, where (a) the phase diagram is obtained after the signal is intercepted and processed for the primary window, and (b) the phase diagram is obtained after the signal is intercepted and processed for the optimized window.
Fig. 5 is an a-line diagram after dispersion compensation of a simulation sample, in which (a) a signal is intercepted for a primary window and processed to obtain a second-order dispersion coefficient for dispersion compensation effect diagram, and (b) a signal is intercepted for an optimized window and processed to obtain a second-order dispersion coefficient for dispersion compensation effect diagram.
Detailed Description
The present invention is further illustrated by the following examples, which should not be construed as limiting the scope of the invention.
The feasibility of the rectangular window function optimization method of windowed FFT in the optical coherence tomography signal processing is analyzed by computer simulation. Referring to fig. 1, which is a diagram of a sample structure a-line with 16 scattering layers, different layers of glass in the sample have dispersive properties, and the thickness of each layer is different, and referring to fig. a, the diagram of the sample a-line without dispersion, signals of each structural layer of the sample are clear, and it can be seen that two signal peaks of a layer with a thin thickness are very close to each other, and two signal peaks of a layer with a relatively thicker thickness are far from each other in the sample. And (b) is a sample A-line graph when dispersion is introduced, and it can be seen that due to dispersion effect, signals of each structural layer of the sample are broadened, and structural layers at certain depths are aliased and cannot be distinguished. When the sample is simulated to be non-dispersive, the interference signal is added with the primary rectangular window intercepted signal and processed to obtain phase errors with different depths, as shown in (a) in figure (2), wherein different line types represent the phase errors of the sample at different depths, and it can be seen that the windowed FFT introduces errors during spectral analysis of each depth, and the errors are horizontal fluctuation within a range of 5 rad. This is because spectral leakage and side lobe effects are unavoidable. The phase error due to improper selection of the window length and center position of the primary rectangular window is large at the two depths indicated by the solid and dashed lines, with the maximum error increasing with the wavenumber to nearly 35 rad. The signal is extracted by the window function optimization method, and referring to (b) in fig. 2, the phase errors at the two depths are greatly reduced, the errors are respectively reduced to be near 0.3rad and 2rad, and the fluctuation is within the range of 5rad in the normal range. And simulating the functional relation between the phase and the wave number when the dispersion exists under the condition of introducing the dispersion. Theoretically, the phase error is quadratic to the wave number, some fluctuation is introduced by the spectrum leakage and the side lobe effect, and referring to fig. 3, it can be seen that the phase error becomes larger with the increase of the wave number due to the introduction of dispersion, the phase error increases to 26-42rad with the wave number under a proper rectangular window, but the error is increased by improper selection of the window length and the center position of the primary rectangular window, as shown by the solid line, the dotted line and the dashed line in the figure, and the phase at these depths is greatly deviated from the actual phase. The specific steps of optimizing the primary window and extracting the second-order dispersion coefficient are as follows:
(1) removing the background of the original interference signal, and performing inverse Fourier transform to obtain an A-line signal;
(2) adding a primary rectangular window to a signal at any depth in the A-line signal to intercept the signal, and performing Fourier transform on the intercepted signal to obtain a phase angle to obtain a primary phase corresponding to the depth signal;
(3) according to the functional relation between the phase and the wave number during dispersion, the primary phase is taken as a target function, a quadratic polynomial of the primary phase and the wave number is obtained by fitting with a least square method, and the standard error of a fitting result is calculated;
(4) changing the window length and the central position of the primary rectangular window to obtain a current rectangular window; intercepting a signal in the depth range by adding a current rectangular window to the A-line signal, performing FFT (fast Fourier transform) on the intercepted signal, then solving a phase angle to obtain a phase corresponding to the depth signal, taking the obtained phase as a target function, fitting by using a least square method to obtain a quadratic polynomial of the phase and a wave number, and calculating a standard error of a fitting result, namely the standard error of the phase corresponding to the rectangular window; then, continuously changing the window length and the central position of the rectangular window, intercepting signals in the depth range, and obtaining standard errors of phases corresponding to different rectangular windows after the rectangular windows traverse all set conditions;
(5) the rectangular window corresponding to the minimum standard error value obtained in the step (4) is the optimal rectangular window, and the optimal rectangular window is utilized to obtain the final phase corresponding to the depth signal;
(6) calculating a corresponding second-order dispersion coefficient according to the final phase of the depth signal;
(7) repeating the steps (2), (3), (4), (5) and (6) for signals with different depths in the A-line signal to obtain second-order dispersion coefficients corresponding to the signals with different depths, and calculating phase deviation caused by dispersion; subtracting the deviation value from the phase corresponding to the depth to obtain the phase after dispersion compensation;
(8) and reconstructing a dispersion-compensated frequency domain interference signal by using each depth-compensated phase.
Under the condition of introducing dispersion, FFT calculation is carried out on signals intercepted by a primary window and an optimized window to obtain a phase which changes along with the depth, and theoretically, the phase is obtained by
Knowing that the phase and the depth are linear, referring to fig. 4, it can be seen that the phase obtained by the primary window has errors at the depths of Z1, Z2, Z3 and Z4, and the phase extracted by the optimization window is found to be more accurate compared with fig. a and b. The second-order dispersion coefficient is calculated from the phase extracted from the primary window and the optimized window for dispersion compensation, see fig. (5) The method can find that the second-order dispersion coefficients of all depths obtained by intercepting signals by a primary window are inaccurate in the corresponding four layers, so that the dispersion compensation effect is poor or even no effect is achieved, the phase error obtained by intercepting the signals by a window function optimization method is small, and the phase of each depth of the sample can be accurately restored, so that the extracted second-order dispersion coefficients changing along with the depth are more accurate, the dispersion compensation effect of the sample on all structural layers is good, and each structural layer can be distinguished.
The above description is only one specific embodiment of the present invention, and the embodiment is only used to illustrate the technical solution of the present invention and not to limit the present invention. The technical solutions available to those skilled in the art through logical analysis, reasoning or limited experiments according to the concepts of the present invention are all within the scope of the present invention.
Claims (4)
1. An optical coherence tomography signal processing method based on rectangular window function optimization is characterized by comprising the following steps:
firstly, a frequency domain optical coherence tomography system is utilized to scan a sample, and a photoelectric detection array of the system records an original interference signal of the sample and transmits the signal to a computer;
secondly, performing inverse Fourier transform on the original interference signal after background removal to obtain an A-line signal;
thirdly, adding a primary rectangular window to a signal in any depth in the A-line signal to intercept the signal, and after Fourier transform is carried out on the intercepted signal, calculating a phase angle to obtain a primary phase corresponding to the depth signal;
fourthly, according to the functional relation between the phase and the wave number during dispersion, the primary phase is taken as a target function, quadratic polynomial of the primary phase and the wave number is obtained by fitting with a least square method, and the standard error of the fitting result is calculated;
changing the window length and the central position of the primary rectangular window to obtain the current rectangular window; intercepting a signal in the depth range by adding a current rectangular window to the A-line signal, performing FFT (fast Fourier transform) on the intercepted signal, then solving a phase angle to obtain a phase corresponding to the depth signal, taking the obtained phase as a target function, fitting by using a least square method to obtain a quadratic polynomial of the phase and a wave number, and calculating a standard error of a fitting result, namely the standard error of the phase corresponding to the rectangular window; then, continuously changing the window length and the central position of the rectangular window, intercepting signals in the depth range, and obtaining standard errors of phases corresponding to different rectangular windows after the rectangular windows traverse all set conditions;
sixthly, the rectangular window corresponding to the minimum standard error value obtained in the fifth step is the optimal rectangular window, and the optimal rectangular window is utilized to obtain the final phase corresponding to the depth signal;
seventhly, calculating a corresponding second-order dispersion coefficient according to the final phase of the depth signal;
sixthly, repeating the step of the signals with different depths in the A-line signal to obtain second-order dispersion coefficients corresponding to the signals with different depths, and calculating phase deviation caused by dispersion; subtracting the deviation value from the phase corresponding to the depth to obtain the phase after dispersion compensation;
and ninthly, reconstructing the frequency domain interference signal after dispersion compensation by utilizing the phase after each depth compensation.
2. The optical coherence tomography signal processing method based on rectangular window function optimization of claim 1, wherein the method for selecting the window length and the center position of the primary rectangular window in the third step is as follows:
when no sample exists, the average amplitude of the A-line signal is the mean value of noise;
when a sample is set, the maximum value of a signal with a certain depth in the A-line signal takes the depth of the maximum value as the central position of the primary rectangular window, and the window is expanded towards two sides until the signal intensity of the two sides is equal to the mean value of noise, and at the moment, the window length of the primary rectangular window is the window length of the depth signal.
3. The optical coherence tomography signal processing method based on rectangular window function optimization of claim 1, wherein the method for changing the window length and center position of the rectangular window in step (v) is:
taking the unit length of the A-line signal as a step length, moving the initial position of the primary rectangular window to the right by 1 step length, sequentially moving the tail end position of the window to the left and the right from 1 step length to N step lengths to obtain 2N rectangular windows in total, intercepting an interference signal under each window to perform Fourier transform to extract a phase, and repeating the operation until the initial position of the window moves to the right by N step lengths; similarly, the initial position of the primary rectangular window is moved by 1 step left, the end position of the window is sequentially moved from 1 step left to N step right to obtain 2N rectangular windows, interference signals under each window are intercepted, Fourier transform is carried out to extract the phase, the process of moving the window initial position by N steps left is repeated until the end, the phase corresponding to the depth signal is obtained when all different rectangular windows intercept signals, and N is 15-30.
4. The method for processing an optical coherence tomography signal based on rectangular window function optimization of claim 1, wherein the step (c) is to calculate a second-order dispersion coefficient corresponding to a depth, specifically:
step 7.1 Fourier transform is carried out on the interference signal at the depth to obtain final phase information
Wherein k represents a wave number,. DELTA.znIs the optical path difference, k, of the sample n-th layer with respect to the reference arm mirror0Number of waves, n, corresponding to the central wavelength of the light sourcenIs the effective refractive index at the nth layer of the sample, ng,nIs the effective group index at the nth layer of the sample, a2Is the second order dispersion compensation coefficient at the nth layer of the sample.
Step 7.2 with (k-k)0) As independent variable, pair
And performing multiple term numerical fitting to obtain a quadratic phase term, namely a second-order dispersion coefficient corresponding to the signal at the depth.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010817011.4A CN111879730B (en) | 2020-08-14 | 2020-08-14 | Optical coherence tomography signal processing method based on rectangular window function optimization |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010817011.4A CN111879730B (en) | 2020-08-14 | 2020-08-14 | Optical coherence tomography signal processing method based on rectangular window function optimization |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111879730A true CN111879730A (en) | 2020-11-03 |
| CN111879730B CN111879730B (en) | 2022-08-30 |
Family
ID=73203822
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010817011.4A Active CN111879730B (en) | 2020-08-14 | 2020-08-14 | Optical coherence tomography signal processing method based on rectangular window function optimization |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111879730B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112683848A (en) * | 2020-12-21 | 2021-04-20 | 中国科学院上海光学精密机械研究所 | Chromatic dispersion compensation method for optical coherence tomography system |
| CN113984713A (en) * | 2021-10-21 | 2022-01-28 | 电子科技大学 | Single-line scanning-based optical coherence tomography chromatic dispersion mismatch detection and compensation method |
| CN115002344A (en) * | 2022-05-11 | 2022-09-02 | 昆山丘钛微电子科技股份有限公司 | Error surface fitting method, device, equipment and medium |
| CN116849626A (en) * | 2023-09-04 | 2023-10-10 | 北京理工大学 | Phase modulation type optical coherence tomography system based on optical calculation |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102332956A (en) * | 2011-08-23 | 2012-01-25 | 天津大学 | Dispersion compensation method for broadband light source |
| CN102628799A (en) * | 2012-04-25 | 2012-08-08 | 中国科学院上海光学精密机械研究所 | Method and system of time-domain optical coherence tomography without depth scan |
| CN104771144A (en) * | 2015-04-13 | 2015-07-15 | 中国科学院上海光学精密机械研究所 | Chromatic dispersion compensation method of FD-OCT (Fourier-Domain Optical Coherence Tomography) system |
| CN107661089A (en) * | 2017-09-19 | 2018-02-06 | 北京工业大学 | A kind of domain optical coherence chromatographs continuous dispersion compensation imaging method and system |
-
2020
- 2020-08-14 CN CN202010817011.4A patent/CN111879730B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102332956A (en) * | 2011-08-23 | 2012-01-25 | 天津大学 | Dispersion compensation method for broadband light source |
| CN102628799A (en) * | 2012-04-25 | 2012-08-08 | 中国科学院上海光学精密机械研究所 | Method and system of time-domain optical coherence tomography without depth scan |
| CN104771144A (en) * | 2015-04-13 | 2015-07-15 | 中国科学院上海光学精密机械研究所 | Chromatic dispersion compensation method of FD-OCT (Fourier-Domain Optical Coherence Tomography) system |
| CN107661089A (en) * | 2017-09-19 | 2018-02-06 | 北京工业大学 | A kind of domain optical coherence chromatographs continuous dispersion compensation imaging method and system |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112683848A (en) * | 2020-12-21 | 2021-04-20 | 中国科学院上海光学精密机械研究所 | Chromatic dispersion compensation method for optical coherence tomography system |
| CN112683848B (en) * | 2020-12-21 | 2022-09-02 | 中国科学院上海光学精密机械研究所 | Chromatic dispersion compensation method for optical coherence tomography system |
| CN113984713A (en) * | 2021-10-21 | 2022-01-28 | 电子科技大学 | Single-line scanning-based optical coherence tomography chromatic dispersion mismatch detection and compensation method |
| CN115002344A (en) * | 2022-05-11 | 2022-09-02 | 昆山丘钛微电子科技股份有限公司 | Error surface fitting method, device, equipment and medium |
| CN115002344B (en) * | 2022-05-11 | 2024-04-19 | 昆山丘钛微电子科技股份有限公司 | Error curved surface fitting method, device, equipment and medium |
| CN116849626A (en) * | 2023-09-04 | 2023-10-10 | 北京理工大学 | Phase modulation type optical coherence tomography system based on optical calculation |
| CN116849626B (en) * | 2023-09-04 | 2024-01-16 | 北京理工大学 | Phase modulation type optical coherence tomography system based on optical calculation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111879730B (en) | 2022-08-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111879730B (en) | Optical coherence tomography signal processing method based on rectangular window function optimization | |
| CN112597947B (en) | Dispersion compensation method based on Fourier domain optical coherence tomography technology | |
| CN112683848B (en) | Chromatic dispersion compensation method for optical coherence tomography system | |
| CN108872153B (en) | Method for measuring optical uniformity of parallel flat plate based on non-uniform Fourier transform | |
| CN112097678B (en) | Multi-surface shape measurement method based on frequency blind estimation | |
| CN111239736B (en) | Single-baseline-based surface elevation correction method, device, equipment and storage medium | |
| CN104655032A (en) | High-precision distance measurement system and method based on orthogonal chromatic dispersion spectral domain interferometer | |
| JPH05281048A (en) | Method for evaluating spatial wave head by relationship of intensity | |
| Zhang et al. | Anti-noise frequency estimation performance of Hanning-windowed energy centrobaric method for optical coherence velocimeter | |
| WO2022233110A1 (en) | Stepped superposition type fourier transform differential method | |
| CN114894308A (en) | Spectrometer calibration method and system based on low coherence interference | |
| CN114022583A (en) | OCT image reconstruction method based on non-uniform discrete Fourier transform | |
| CN116336936B (en) | Fourier domain OCT system dispersion compensation method and system based on wavelength modulation | |
| CN103822717A (en) | Method for enhancing broadband interference spectral resolution and spectrograph | |
| CN110864808B (en) | Fourier transform spectrum detection method based on high-speed sampling | |
| Cai et al. | A new adaptive multiresolution noise-filtering approach for SAR interferometric phase images | |
| CN116879894A (en) | Phase unwrapping method and system for large gradient deformation area of mining area | |
| CN116091368A (en) | Dispersion compensation method, device, electronic equipment and storage medium | |
| CN114322833B (en) | White light scanning interference three-dimensional reconstruction method based on pseudo Wigner-Ville distribution | |
| Barajas et al. | Towards an on-chip signal processing solution for the online calibration of SS-OCT systems | |
| CN114565541A (en) | InSAR filtering method based on CZT spectrum refinement | |
| CN204575033U (en) | Based on the high precision distance measurement system of cross dispersion spectral domain interferometer | |
| CN112034457B (en) | Multi-baseline elevation interference phase estimation method based on interference fringe direction | |
| CN109580547B (en) | Segmented four-step phase shifting method for removing complex conjugate mirror image by frequency domain optical coherence tomography | |
| CN116818717B (en) | Compensation method and system for wavelength calibration error in spectral domain optical coherence tomography |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |