Sweep frequency OCT imaging system phase correction method based on GPU acceleration

Abstract

The invention discloses a sweep frequency OCT imaging system phase correction method based on GPU acceleration, and belongs to the technical field of signal processing. The method comprises the steps that an SS-OCT system is built, and an MZI reference signal and an OCT interference signal are collected; carrying out local upsampling on the MZI signals, and calculating the sub-pixel translation amount of each column of signals through threshold detection and reduction operation; correcting the nonlinear error of the OCT signal by using frequency domain phase compensation; wave number domain resampling is realized based on reference column Hilbert transform; according to the method, the processes of up-sampling, frequency domain translation and phase unwrapping are optimized through CUDA parallel computing, redundant operation is reduced in combination with a local computing strategy, the single-frame data processing time is greatly shortened, the scanning rate is increased to 2.48 times that of a traditional method, the improvement effect is obvious, and the data processing requirement of higher-rate SS-OCT can be met.

Description

Technical Field

[0001] The present invention relates to a phase correction method for a swept-frequency OCT imaging system based on GPU acceleration, and belongs to the technical field of signal processing. Background Art

[0002] Optical coherence tomography (OCT) is a high-resolution, non-contact optical imaging technology that can capture the three-dimensional structure of objects. It is widely used in biomedical imaging fields such as ophthalmology (retina, cornea), cardiovascular, and skin. Swept-source OCT (SS-OCT) uses a tunable laser as its light source, enabling deeper imaging depth, faster scanning speeds, and higher signal-to-noise ratios than previous generations of technology.

[0003] SS-OCT signals require a series of signal processing steps before they can be converted into diagnostic images. The most important step is wavenumber correction. However, compared to previous generations of technology, SS-OCT systems introduce greater phase noise, which affects the SS-OCT imaging denoising results and subsequent applications such as SS-OCT functional imaging that utilize phase information (such as blood flow imaging based on SS-OCT systems). Therefore, this is a major difficulty in real-time imaging using SS-OCT systems.

[0004] Currently, different software and hardware methods have been used to solve this problem. Choi WJ. et al. used a method to detect specific wavelengths by adding a fiber Bragg grating in front of a balanced detector. This method increases the complexity of the hardware. At the same time, due to the use of hardware wavenumber sampling, it can only correct offsets of integer multiples of 2π, which is not conducive to the expansion of imaging depth (Choi WJ, Potsaid B, Jayaraman V, et al. Phase-sensitive swept source OCT imaging of the human retina with a VCSEL light source [J]. Optics Letters, 2013, 38 (3). DOI: 10.1364 / ol.38.000338.). Braaf B. et al. used another light path to construct a Mach-Zehnder interferometer (MZI) and used Hilbert transform to analyze the phase space of the MZI signal. However, this method truncated the phase interval and applied a one-to-many Hilbert transform to the signal, which greatly increased the amount of computation (Braaf B et al. Phase-stabilized optical frequency domain imaging at 1-m for the measurement of blood flow in the human choroid [J]. Optics Express, 2011, 19(21): 20886-20903. DOI: 10.1364 / OE.19.020886.). Shangguan Ziwei et al. proposed to use the cross-correlation method in the Hilbert transform domain to improve the rate. However, they did not solve the problem of the computational complexity of fast Fourier transform (FFT) and phase unwrapping required for Hilbert transform of each column of MZI signal. At the same time, this method cannot correct the phase offset at the sub-pixel level (Shangguan Ziwei, Shen Yi, Li Peng, et al. Study on wavenumber correction and phase measurement of swept-frequency optical coherence tomography system [J]. Acta Physica Sinica, 2016(3):6.DOI:10.7498 / aps.65.034201.).Fan Jinyu et al. improved the cross-correlation method in the previous method and proposed a method of using time domain cross-correlation instead of taking Hilbert transform and then performing cross-correlation on the Mzi signal. However, in order to correct the sub-pixel phase offset, this method uses a frequency domain upsampling method, which also has a large amount of calculation and is not conducive to the real-time signal processing of higher-speed SS-OCT (Fan Jinyu, Gao Feng, Kong Wen, et al. Full spectrum resampling method for multi-faceted rotating mirror laser scanning optical coherence tomography system [J]. Acta Physica Sinica, 2017, 66(11):10.DOI:10.7498 / aps.66.114204.).

[0005] In summary, the various SS-OCT signal resampling and phase correction schemes currently proposed have the problems of high system cost and large computational complexity, which is not conducive to the construction of SS-OCT systems with faster scanning rates, deeper imaging depths, and higher signal-to-noise ratios. Summary of the Invention

[0006] In order to reduce system cost and computational complexity, increase scanning speed, and build an SS-OCT system with a deeper imaging depth and higher signal-to-noise ratio, the present invention provides a phase correction method for a swept-frequency OCT imaging system based on GPU acceleration. The technical solution is as follows:

[0007] Step 1: Build the SS-OCT hardware system, including the spectrometer module, MZI optical path, sample optical path, and reference optical path. The swept laser source is split by the spectrometer module and then input into the MZI optical path, sample optical path, and reference optical path respectively. The MZI optical path signal and the sample / reference optical path signal are respectively received by the balanced detector and transmitted to the host computer via the data acquisition card.

[0008] Step 2: For each frame of OCT image data, select the first N points of each column of MZI reference signal and perform linear upsampling interpolation processing;

[0009] Step 3: Set the MZI reference signal amplitude threshold Th and obtain the coordinate index of each column where the signal reaches the threshold;

[0010] Step 4: Calculate the sub-pixel shift of each column signal relative to the MZI reference signal:

[0011]

[0012] Among them, m k Indicates the coordinate index of the kth column that reaches the threshold, m reference Indicates the coordinate index where the MZI reference signal reaches the threshold, L indicates the upsampling multiple, Δt k is the calculated sub-pixel translation;

[0013] Step 5: Perform translation compensation on each column of raw OCT signals in the frequency domain:

[0014]

[0015] in, represents the inverse Fourier transform, y k [n] is the kth column of original OCT signal, y′ k [n] is the kth column of original OCT signal after translation, n is the signal position index, N ori is the length of each column of the original OCT signal;

[0016] Step 6: Perform Hilbert transform and phase unwrapping on the MZI reference signal to obtain the calibration phase signal;

[0017] Step 7: performing wavenumber domain resampling based on the inverse mapping relationship between the calibration phase signal and the signal index n to obtain a corrected signal;

[0018] Step 8: Perform Fourier transform and post-processing on the corrected signal to generate an OCT image.

[0019] Optionally, the step 2 uses a linear upsampling interpolation:

[0020]

[0021] Where x[n] is the value of the original MZI reference signal, L is the upsampling multiple, x[n′] is the upsampled MZI reference signal, and n′, n are signal position indices.

[0022] Optionally, the expression for obtaining the calibrated phase signal in step 6 is:

[0023]

[0024] in represents Hilbert transform, i represents imaginary number, represents the MZI signal after Hilbert transform, Im represents the real part, Re represents the imaginary part, unwrap represents unwrapping, Δφ n represents the final Hilbert phase.

[0025] Optionally, the step 7 uses the Hilbert phase Δφ n Calibrate in the wavenumber domain according to n and Δφ n The relationship between the two can be inferred inversely that n Interval, the index that needs to be transformed Here, Δφ n Divide into equally spaced k1,…k N , get the new index Using the calculated The relationship between the original signal is interpolated:

[0026]

[0027] f -1 Represents n and Δφ n The inverse function between Indicates the interpolation method, y final [n] is the signal for final processing.

[0028] A second object of the present invention is to provide a phase correction system for a swept frequency OCT imaging system, the system comprising:

[0029] Splitting module, MZI optical path, sample optical path and reference optical path;

[0030] Two balanced detectors, connected to the MZI optical path and the sample / reference optical path respectively;

[0031] Data acquisition card, used for synchronously collecting the signals output by the balanced detector;

[0032] A host computer is configured with a CUDA-supported GPU for executing a phase correction method, the method comprising:

[0033] Step 1: Build the SS-OCT hardware system, including the spectrometer module, MZI optical path, sample optical path, and reference optical path. The swept laser source is split by the spectrometer module and then input into the MZI optical path, sample optical path, and reference optical path respectively. The MZI optical path signal and the sample / reference optical path signal are respectively received by the balanced detector and transmitted to the host computer via the data acquisition card.

[0034] Step 2: For each frame of OCT image data, select the first N points of each column of MZI reference signal and perform linear upsampling interpolation processing;

[0035] Step 3: Set the MZI reference signal amplitude threshold Th and obtain the coordinate index of each column where the signal reaches the threshold;

[0036] Step 4: Calculate the sub-pixel shift of each column signal relative to the MZI reference signal:

[0037]

[0038] Among them, m k Indicates the coordinate index of the kth column that reaches the threshold, m reference Indicates the coordinate index where the MZI reference signal reaches the threshold, L indicates the upsampling multiple, Δt k is the calculated sub-pixel translation;

[0039] Step 5: Perform translation compensation on each column of OCT signals in the frequency domain:

[0040]

[0041] in, represents the inverse Fourier transform, y k [n] is the kth column of original OCT signal, y′ k [n] is the kth column of original OCT signal after translation, n is the signal position index, N ori is the length of each column of the original OCT signal;

[0042] Step 6: Perform Hilbert transform and phase unwrapping on the MZI reference signal to obtain the calibration phase signal;

[0043] Step 7: performing wavenumber domain resampling based on the inverse mapping relationship between the calibration phase signal and the signal index n to obtain a corrected signal;

[0044] Step 8: Perform Fourier transform and post-processing on the corrected signal to generate an OCT image.

[0045] Optionally, the step 2 uses a linear upsampling interpolation:

[0046]

[0047] Where x[n] is the value of the original MZI reference signal, L is the upsampling multiple, x[n′] is the upsampled MZI reference signal, and n′, n are signal position indices.

[0048] Optionally, the expression for obtaining the calibrated phase signal in step 6 is:

[0049]

[0050] in represents Hilbert transform, i represents imaginary number, represents the MZI signal after Hilbert transform, Im represents the real part, Re represents the imaginary part, unwrap represents unwrapping, Δφ n represents the final Hilbert phase.

[0051] Optionally, the step 7 uses the Hilbert phase Δφ n Calibrate in the wavenumber domain according to n and Δφ n The relationship between the two can be inferred inversely that n Interval, the index that needs to be transformed Here, Δφ n Divide into equally spaced k1,…k N, get the new index Using the calculated The relationship between the original signal is interpolated:

[0052]

[0053] f -1 Represents n and Δφ n The inverse function between Indicates the interpolation method, y final [n] is the signal for final processing.

[0054] The third object of the present invention is to provide an electronic device, comprising a memory and a GPU; the memory is used to store a computer program; and the GPU is used to implement any of the methods described above when executing the computer program.

[0055] A fourth object of the present invention is to provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a GPU, implements any of the methods described above.

[0056] The beneficial effects of the present invention are:

[0057] The present invention uses GPU parallel acceleration, and all key steps are based on CUDA optimization. Compared with traditional CPU serial calculation, the calculation time is significantly reduced. Only the first N points of the MZI reference signal are upsampled (rather than the entire column of data), further reducing redundant calculations. In addition, the threshold coordinates are quickly calculated through reduction operations, reducing the complexity of comparative calculations. Experiments have shown that the present invention has a faster processing rate, with the number of processed frames per second reaching more than 2.48 times that of existing methods. The improvement effect is obvious and can meet the data processing requirements of higher-speed SS-OCT. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0059] Figure 1 It is a structural block diagram of the swept-frequency OCT imaging phase correction system of the present invention.

[0060] Figure 2 This is a flow chart of the swept-frequency OCT imaging phase correction method of the present invention.

[0061] Figure 3This is a schematic diagram of the MZI reference signal sampling offset provided in the first embodiment of the present invention.

[0062] Figure 4 This is a schematic diagram of a corrected signal provided in the first embodiment of the present invention.

[0063] Figure 5 This is the corrected phase difference diagram provided by the first embodiment of the present invention.

[0064] Figure 6 1 is a result diagram before and after signal processing provided by Example 1 of the present invention. DETAILED DESCRIPTION

[0065] To make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0066] Example 1:

[0067] This embodiment provides a phase correction method for a swept-frequency OCT imaging system based on GPU acceleration, including the following steps:

[0068] Step 1: First build Figure 1 The SS-OCT hardware system shown in the figure adopts the classic SS-OCT system solution in this embodiment, including: a spectrometer module, an MZI optical path, a sample optical path, and a reference optical path. The swept laser source is divided into the MZI optical path, the sample optical path, and the reference optical path by the spectrometer module. The interference signal of the MZI optical path is received by a balanced detector and input into a data acquisition card, and then transmitted to the host computer. The sample optical path and the reference optical path are received by another balanced detector and input into the data acquisition card, and then transmitted to the host computer.

[0069] The OCT reference signal (MZI reference signal) collected in the MZI optical path must not exceed 1 / 10 of the acquisition card sampling bandwidth to reduce the impact of noise. The collected MZI reference signal is as follows: Figure 3 shown.

[0070] Step 2: Sample the MZI reference signal.

[0071] For each OCT image frame, assuming the number of image columns is M, the MZI reference signal corresponding to each column can be obtained. The first N points in each column of the MZI reference signal are selected for upsampling (to save efficiency, there is no need to upsample the entire column of reference signals). To optimize the rate as much as possible, this embodiment uses a linear upsampling interpolation:

[0072]

[0073] Where x[n] is the value of the original MZI reference signal, L is the upsampling factor, x[n′] is the upsampled signal, and n′ and n are the indices of the corresponding positions. Upsampling can be optimized and accelerated using texture memory in CUDA-enabled graphics cards. Other upsampling methods, including but not limited to cubic spline interpolation, high-order polynomial interpolation, and frequency-domain Fourier interpolation, can achieve higher accuracy, but with a corresponding loss in time efficiency.

[0074] Step 3: Set the threshold Th of the MZI reference signal amplitude.

[0075] The wavelength distribution of the swept laser is relatively stable and does not cause deformation of the interference signal. Taking advantage of this, we manually set the amplitude threshold Th and find the coordinate index where each individual MZI signal reaches the threshold:

[0076] m={k|x[k]=Th}

[0077] In order to simplify the difficulty, the Th value is set here as the maximum value of the column signal. Using a graphics card that supports CUDA and the reduction operation method can greatly reduce the time required for this step of calculation.

[0078] Step 4: Use the obtained threshold coordinate index corresponding value {m k , k=1,2,…M}, and subtract it from the MZI reference signal to obtain the sub-pixel shift of the signal relative to the MZI reference signal:

[0079]

[0080] where Δt k The calculated translation amount compared to the MZI reference signal, m reference is the coordinate index where the MZI reference signal reaches the threshold. CUDA can maximize parallel computing to obtain the required value.

[0081] Step 5: Use the translation Δt obtained in the previous step k , each column of OCT signals is shifted. In particular, in order to accurately shift the signals, this embodiment adopts a frequency domain method:

[0082]

[0083] in represents the Fourier transform, represents the inverse Fourier transform, * represents the dot product of the discrete sequence, y k [n] is the kth column of original discrete OCT signal, y′ k [n] is the kth column of original OCT signal after translation, n is the index of the signal in each column, N oriis the length of each column of the original signal. Using a CUDA-supported graphics card and the cufft library for coding can significantly reduce the required computation time.

[0084] Step 6: After the signal is shifted, the conventional OCT signal resampling correction process can be performed to interpolate the signal sampled in the time domain back to the OCT signal sampled in the wavenumber domain. First, the MZI reference signal is Hilbert transformed and phase unwrapped to obtain the phase signal for calibration:

[0085]

[0086]

[0087] in represents Hilbert transform, i represents imaginary number, represents the MZI signal after Hilbert transform, Im represents the real part, Re represents the imaginary part, unwrap represents unwrapping, and removes the stacking caused by the 2π period of the phase, Δφ n represents the final Hilbert phase.

[0088] Step 7: Use the obtained Hilbert phase signal to calibrate the wavenumber domain according to n and Δφ n The relationship between the two can be inferred inversely that n Interval, the index that needs to be transformed Here, Δφ n Divide into equally spaced k1,…k N , get the new index Using the calculated The relationship between the original signal is interpolated:

[0089]

[0090] f -1 Represents n and Δφ n The inverse function between Represents the interpolation method, and the corresponding relationship can be obtained by polynomial fitting or cubic spline fitting. final [n] is the signal obtained for final processing.

[0091] Step 8: Finally, the obtained y final [n] Performing Fourier transform, taking the logarithmic transform domain, and normalizing can obtain the components of each column in the OCT image for display. The entire process can be run on the GPU using a CUDA-supported graphics card.

[0092] In order to further illustrate the effects of the present invention, a series of related experiments and analytical comparisons were carried out.

[0093] The experimental conditions are as follows:

[0094] We built an SS-OCT system with a 100 kHz sweep rate. We used an acquisition card with a 1 GHz sampling rate and acquired OCT images of 1024 x 8192 pixels. Based on the sweep rate, we calculated that each frame took approximately 10 milliseconds to acquire. The computing platform used a 64-bit Windows operating system, an Intel i9 14900kF CPU, and an Nvidia GeForce RTX 4070 Super GPU with 12GB of video memory.

[0095] The collected signals are as follows Figure 3 As shown in FIG, it can be seen that there is a time delay of non-integer sampling intervals between different signal sequences, which is the sub-pixel sampling error described in the present invention. The signal corrected by the phase correction method of the present invention is shown in FIG. Figure 4 As shown in the figure, the alignment is basically achieved in the time domain. From the phase domain, the following is analyzed: Figure 5 As shown in Figure 3, the phase error is greatly reduced, with the standard deviation dropping from 0.199 rad to 0.02 rad, which is sufficient for common OCT signal processing.

[0096] Finally, the theoretical computational complexity and actual running time of the algorithm mentioned in the present invention and the paper "Fan Jinyu, Gao Feng, Kong Wen, et al. Full spectrum resampling method for multi-faceted rotating mirror laser sweep optical coherence tomography system [J]. Acta Physica Sinica, 2017, 66(11):10.DOI:10.7498 / aps.66.114204." are analyzed. Assuming that the upsampling multiple is L, the size of an image is M*N, and the length of the intercepted signal used for upsampling is K, if only the number of time-consuming multiplication operations is considered, as shown in Table 1:

[0097] Table 1 Comparative analysis of the two algorithms

[0098]

[0099]

[0100] From Table 1, it can be seen that the theoretical amount of computation required by the present invention is smaller, and the actual processing process has a faster processing rate, with the number of processing frames per second reaching more than 2.48 times, which is a significant improvement. This method can meet the data processing requirements of higher-speed SS-OCT.

[0101] Example 2:

[0102] This embodiment provides a phase correction system for a swept-frequency OCT imaging system, including:

[0103] Splitting module, MZI optical path, sample optical path and reference optical path;

[0104] Two balanced detectors, connected to the MZI optical path and the sample / reference optical path respectively;

[0105] Data acquisition card, used for synchronously collecting the signals output by the balanced detector;

[0106] A host computer is configured with a CUDA-supported GPU for executing a phase correction method, the method comprising:

[0107] Step 1: Build the SS-OCT hardware system, including the spectrometer module, MZI optical path, sample optical path, and reference optical path. The swept laser source is split by the spectrometer module and then input into the MZI optical path, sample optical path, and reference optical path respectively. The MZI optical path signal and the sample / reference optical path signal are respectively received by the balanced detector and transmitted to the host computer via the data acquisition card.

[0108] Step 2: For each frame of OCT image data, select the first N points of each column of MZI reference signal and perform linear upsampling interpolation processing;

[0109] Step 3: Set the MZI reference signal amplitude threshold Th and obtain the coordinate index of each column where the signal reaches the threshold;

[0110] Step 4: Calculate the sub-pixel shift of each column signal relative to the MZI reference signal:

[0111]

[0112] Among them, m k Indicates the coordinate index of the kth column that reaches the threshold, m reference Indicates the coordinate index where the MZI reference signal reaches the threshold, and L represents the upsampling multiple;

[0113] Step 5: Perform translation compensation on each column of raw OCT signals in the frequency domain:

[0114]

[0115] in, represents the inverse Fourier transform, y k [n] is the kth column of original OCT signal, y′ k [n] is the kth column of original OCT signal after translation, n is the signal position index, N ori is the length of each column of the original OCT signal;

[0116] Step 6: Perform Hilbert transform and phase unwrapping on the MZI reference signal to obtain the calibration phase signal;

[0117] Step 7: performing wavenumber domain resampling based on the inverse mapping relationship between the calibration phase signal and the signal index n to obtain a corrected signal;

[0118] Step 8: Perform Fourier transform and post-processing on the corrected signal to generate an OCT image.

[0119] Some steps in the embodiments of the present invention may be implemented using software, and the corresponding software program may be stored in a readable storage medium, such as a CD or a hard disk.

[0120] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A phase correction method for a swept frequency OCT imaging system, characterized in that: The method is accelerated on a GPU through CUDA parallel computing, and includes: Step 1: Build the SS-OCT hardware system, including the spectrometer module, MZI optical path, sample optical path, and reference optical path. The swept laser source is split by the spectrometer module and then input into the MZI optical path, sample optical path, and reference optical path respectively. The MZI optical path signal and the sample / reference optical path signal are respectively received by the balanced detector and transmitted to the host computer via the data acquisition card. Step 2: For each frame of OCT image data, select the first N points of each column of MZI reference signal and perform linear upsampling interpolation processing; Step 3: Set the MZI reference signal amplitude threshold Th and obtain the coordinate index of each column where the signal reaches the threshold; Step 4: Calculate the sub-pixel shift of each column signal relative to the MZI reference signal: Among them, m k Indicates the coordinate index of the kth column that reaches the threshold, m reference Indicates the coordinate index of the MZI reference signal reaching the threshold, and L represents the upsampling multiple; Step 5: Perform translation compensation on each column of raw OCT signals in the frequency domain: in, represents the inverse Fourier transform, y k [n] is the kth column of original OCT signal, y′ k [n] is the kth column of original OCT signal after translation, n is the signal position index, N ori is the length of each column of the original OCT signal; Step 6: Perform Hilbert transform and phase unwrapping on the MZI reference signal to obtain the calibration phase signal; Step 7: performing wavenumber domain resampling based on the inverse mapping relationship between the calibration phase signal and the signal index n to obtain a corrected signal; Step 8: Perform Fourier transform and post-processing on the corrected signal to generate an OCT image.

2. The phase correction method for a swept frequency OCT imaging system according to claim 1, characterized in that: The step 2 uses a linear upsampling interpolation: Where x[n] is the value of the original MZI reference signal, L is the upsampling multiple, x[n′] is the upsampled MZI reference signal, and n′, n are signal position indices.

3. The phase correction method for a swept frequency OCT imaging system according to claim 1, wherein: The acquisition expression of the calibration phase signal in step 6 is: in represents Hilbert transform, i represents imaginary number, represents the MZI signal after Hilbert transform, Im represents the real part, Re represents the imaginary part, unwrap represents unwrapping, Δφ n represents the final Hilbert phase.

4. The phase correction method for a swept frequency OCT imaging system according to claim 1, wherein: Step 7 uses the Hilbert phase Δφ n Calibrate in the wavenumber domain according to n and Δφ n The relationship between the two can be inferred inversely that n Interval, the index that needs to be transformed Here, Δφ n Divide into equally spaced k1,…k N , get the new index Using the calculated The relationship between the original signal is interpolated: f -1 Represents n and Δφ n The inverse function between Indicates the interpolation method, y final [n] is the signal for final processing.

5. A phase correction system for a swept frequency OCT imaging system, characterized in that: The system comprises: Splitting module, MZI optical path, sample optical path and reference optical path; Two balanced detectors, connected to the MZI optical path and the sample / reference optical path respectively; Data acquisition card, used for synchronously collecting the signals output by the balanced detector; A host computer is configured with a CUDA-supported GPU for executing a phase correction method, the method comprising: Step 1: Build the SS-OCT hardware system, including the spectrometer module, MZI optical path, sample optical path, and reference optical path. The swept laser source is split by the spectrometer module and then input into the MZI optical path, sample optical path, and reference optical path respectively. The MZI optical path signal and the sample / reference optical path signal are respectively received by the balanced detector and transmitted to the host computer via the data acquisition card. Step 2: For each frame of OCT image data, select the first N points of each column of MZI reference signal and perform linear upsampling interpolation processing; Step 3: Set the MZI reference signal amplitude threshold Th and obtain the coordinate index of each column where the signal reaches the threshold; Step 4: Calculate the sub-pixel shift of each column signal relative to the MZI reference signal: Among them, m k Indicates the coordinate index of the kth column that reaches the threshold, m reference Indicates the coordinate index of the MZI reference signal reaching the threshold, L indicates the upsampling multiple, Δt k is the calculated sub-pixel translation; Step 5: Perform translation compensation on each column of raw OCT signals in the frequency domain: in, represents the inverse Fourier transform, y k [n] is the kth column of original OCT signal, y′ k [n] is the kth column of original OCT signal after translation, n is the signal position index, N ori is the length of each column of the original OCT signal; Step 6: Perform Hilbert transform and phase unwrapping on the MZI reference signal to obtain the calibration phase signal; Step 7: performing wavenumber domain resampling based on the inverse mapping relationship between the calibration phase signal and the signal index n to obtain a corrected signal; Step 8: Perform Fourier transform and post-processing on the corrected signal to generate an OCT image.

6. The phase correction system for a swept frequency OCT imaging system according to claim 5, characterized in that: The step 2 uses a linear upsampling interpolation: Where x[n] is the value of the original MZI reference signal, L is the upsampling multiple, x[n′] is the upsampled MZI reference signal, and n′, n are signal position indices.

7. The phase correction system for a swept frequency OCT imaging system according to claim 5, characterized in that: The acquisition expression of the calibration phase signal in step 6 is: in represents Hilbert transform, i represents imaginary number, represents the MZI signal after Hilbert transform, Im represents the real part, Re represents the imaginary part, unwrap represents unwrapping, Δφ n represents the final Hilbert phase.

8. The phase correction system for a swept frequency OCT imaging system according to claim 5, characterized in that: Step 7 uses the Hilbert phase Δφ n Calibrate in the wavenumber domain according to n and Δφ n The relationship between the two can be inferred inversely that n Interval, the index that needs to be transformed Here, Δφ n Divide into equally spaced k1,…k N , get the new index Using the calculated The relationship between the original signal is interpolated: f -1 Represents n and Δφ n The inverse function between Indicates the interpolation method, y final [n] is the signal for final processing.

9. An electronic device, characterized in that: The method comprises a memory and a GPU; the memory is used to store a computer program; the GPU is used to implement the method according to any one of claims 1 to 4 when executing the computer program.

10. A computer-readable storage medium, characterized in that The storage medium stores a computer program, and when the computer program is executed by the GPU, the method according to any one of claims 1 to 4 is implemented.

Images