Adaptive sub-pixel decorrelation method and system for optical coherence tomography

Abstract

The invention aims to provide a self-adaptive sub-pixel decorrelation method and system for optical coherence tomography. The method comprises the following steps: acquiring interference spectrum data before and after deformation of a sample; obtaining a high-resolution phase difference and a high-resolution phase quality distribution diagram; performing merging processing on the interference spectrum data to obtain a low-resolution phase difference after axial merging, a low-resolution phase mass distribution diagram after axial merging, a low-resolution phase difference after transverse merging and a low-resolution phase mass distribution diagram after transverse merging; adaptively selecting an optimal phase difference result to obtain a final output phase difference; and outputting the final output phase difference diagram. According to the invention, phase decorrelation noise caused by sub-pixel displacement offset can be effectively suppressed; moreover, while noise is suppressed, the spatial resolution of imaging is reserved to the greatest extent, and seamless fusion with the existing pixel-level compensation technology is realized.

Description

Technical Field

[0001] This invention relates to the field of optical imaging and measurement technology, specifically to a method for phase-sensitive optical coherence tomography (PCT). PhS - OCT In this paper, an adaptive method and system are proposed to suppress phase decorrelation noise caused by subpixel displacement and improve the signal-to-noise ratio of full-field phase difference measurement. Background Technology

[0002] Phase-sensitive optical coherence tomography (PCT) PhS - OCT ) is a high-precision optical imaging technology that can realize the measurement of nanoscale displacement, deformation and strain field inside a sample. It has been widely used in biomedical diagnostics (such as blood flow imaging, tissue mechanical property characterization) and non-destructive testing.

[0003] exist PhS - OCT In this method, displacement information is obtained by calculating the phase difference of the interference signals before and after sample deformation. However, when the sample deforms or translates, the scatterers inside it may move across pixels (i.e., the displacement is not an integer number of pixels), causing the signals compared when calculating the phase difference to not originate from the same scatterer, thus introducing random distribution in [- π , π Phase decorrelation noise significantly reduces the signal-to-noise ratio. SNR This can even lead to measurement failure.

[0004] Existing technologies primarily focus on addressing decorrelation issues caused by pixel-level displacement. These methods, through displacement tracking and compensation, can effectively recover phase information lost due to integer pixel displacement, reducing the amplitude of decorrelation noise from... π Reduced to 0.5 π .

[0005] However, existing technologies lack effective solutions to the sub-pixel decorrelation problem. Typically, after pixel-level compensation, in the boundary region where the displacement offset changes, the proportion of the scatterer moving across pixels ( R The value may still be close to 50%, resulting in a decorrelation noise amplitude as high as 0.5. π This causes phase quality issues in these regions ( SNR () decreased significantly.

[0006] Therefore, a method for suppressing phase decorrelation noise caused by subpixel displacement urgently needs to be developed. Summary of the Invention

[0007] The purpose of this invention is to provide an adaptive sub-pixel decorrelation method and system for optical coherence tomography, in order to solve a technical problem in the prior art.

[0008] The technical solution of this invention is: An adaptive sub-pixel decorrelation method for optical coherence tomography includes: Obtain interference spectral data of the sample before and after deformation; High-resolution phase difference and high-resolution phase quality distribution map are obtained using the interference spectral data; and the interference spectral data are merged to obtain low-resolution phase difference and low-resolution phase quality distribution map after axial merging, as well as low-resolution phase difference and low-resolution phase quality distribution map after lateral merging. Based on the high-resolution phase difference and high-resolution phase quality distribution map, the low-resolution phase difference after axial merging, the low-resolution phase quality distribution map after axial merging, the low-resolution phase difference after lateral merging, and the low-resolution phase quality distribution map after lateral merging, the optimal phase difference result is adaptively selected to obtain the final output phase difference. Output the final phase difference diagram.

[0009] The process of acquiring high-resolution phase difference and high-resolution phase quality distribution maps from the interferometric spectral data includes: The complex values ​​of the sample before and after deformation are obtained based on the interference spectral data. OCT The signal was analyzed to obtain a high-resolution phase map. The high-resolution phase difference and high-resolution phase quality distribution map corresponding to the high-resolution phase map are obtained by using a pixel-level decorrelation compensation method.

[0010] The process of obtaining the low-resolution phase difference and the low-resolution phase quality distribution map after axial merging includes: The interferometric spectral data is used to perform axial pixel merging processing to obtain the corresponding axial low-resolution interferometric spectral data; The pixel-level decorrelation compensation method is used to process the axial low-resolution interferometric spectral data to obtain the axially merged low-resolution phase difference and the axially merged low-resolution phase quality distribution map.

[0011] The process of obtaining the low-resolution phase difference after horizontal merging and the low-resolution phase quality distribution map after axial merging includes: The interferometric spectral data is used to perform lateral pixel merging processing to obtain the corresponding lateral low-resolution interferometric spectral data; The pixel-level decorrelation compensation method is used to process the lateral low-resolution interferometric spectral data to obtain the laterally merged low-resolution phase difference and the laterally merged low-resolution phase quality distribution map, respectively.

[0012] The process of adaptively selecting the optimal phase difference result based on the high-resolution phase difference and high-resolution phase quality distribution map, the low-resolution phase difference after axial merging, the low-resolution phase quality distribution map after axial merging, and the low-resolution phase difference and low-resolution phase quality distribution map after lateral merging, to obtain the final output phase difference map includes: The final output phase difference map is generated by adaptively selecting the optimal phase difference result according to the following formula: ; in: Δφ o For high-resolution phase difference; D oT As a judgment factor; Δφ a This represents the low-resolution phase difference after axial merging. D a This is a low-resolution phase quality distribution map after axial merging; Δφ l This represents the low-resolution phase difference after horizontal merging. D l This is a low-resolution phase quality distribution map after horizontal merging.

[0013] The judgment factor D oT for: D oT = Do – Thr ;in, D o A high-resolution phase quality distribution map; Thr The preset threshold; Based on the aforementioned judgment factors D oT Obtain the optimal spatial resolution.

[0014] The judgment factor D oT To obtain the optimal spatial resolution, including: High-resolution phase quality distribution map corresponding to high-resolution phase difference D o With the threshold Thr The results are compared, and the high-resolution phase difference results are retained or the low-resolution phase difference results are used as the final output.

[0015] The high-resolution phase quality distribution map corresponding to the high-resolution phase difference D o With the threshold Thr The comparison is performed, retaining the high-resolution result corresponding to the high-resolution phase difference or using the low-resolution phase difference result as the final output, including: High-resolution phase quality distribution map corresponding to high-resolution phase difference D o With the threshold Thr Comparison: If the high-resolution phase difference corresponds to the high-resolution phase quality distribution map D o Less than the threshold Thr If so, the corresponding high-resolution phase difference is selected as the output for that region; If the high-resolution phase difference corresponds to the high-resolution phase quality distribution map D o Greater than the threshold Thr If the low-resolution phase difference result is selected as the output for that region, then the output for that region will be chosen.

[0016] The selection of low-resolution phase difference results as the output for this region includes: The low-resolution phase quality distribution maps after axial merging and the low-resolution phase quality distribution maps after lateral merging are obtained by using the low-resolution phase difference after axial merging or the low-resolution phase difference after lateral merging, respectively. By comparing the low-resolution phase quality distribution map after axial merging and the low-resolution phase quality distribution map after lateral merging, the lower low-resolution phase difference after axial merging or the lower low-resolution phase difference after lateral merging is taken as the output of this region.

[0017] The beneficial effects of the present invention include at least the following: The method described in this invention can effectively suppress phase decorrelation noise caused by sub-pixel displacement; furthermore, while suppressing noise, it maximizes the preservation of spatial resolution of the image, achieving seamless integration with existing pixel-level compensation techniques and further improving... PhS - OCT Accuracy and dynamic range of full-field phase measurement. Attached Figure Description

[0018] Figure 1 This is a flowchart of the method described in this invention. Detailed Implementation

[0019] The present application will now be further described with reference to the accompanying drawings.

[0020] The purpose of this invention is to overcome the existingPhS - OCT To address the aforementioned shortcomings of pixel-level decorrelation compensation techniques, an adaptive subpixel decorrelation method and system for optical coherence tomography is provided. This system comprises two levels of decorrelation compensation: pixel-level compensation and subpixel-level adaptive compensation. It can simultaneously overcome pixel-level and subpixel-level decorrelation noise, significantly improving the accuracy and robustness of full-field phase measurement.

[0021] like Figure 1 The main steps of the method described in this invention include: S 1. Data Acquisition and Initial Processing: use OCT The system acquires the interference spectrum data of the sample before and after deformation, and calculates the complex values ​​before and after deformation respectively. OCT The signal is then used to extract a high-resolution phase map.

[0022] S1 Pixel-level decorrelation compensation: Applying existing pixel-level decorrelation compensation algorithms, such as those based on the above high-resolution phase map, to the high-resolution phase map is a good approach. D - ratio The registration method yields a high-resolution phase difference map after preliminary compensation. Δφ o and its corresponding phase-mass distribution diagram D o Phase quality distribution diagram D o The value of directly reflects the noise level of the phase.

[0023] S2 Pixel merging ( Binning )deal with: Simultaneously, the original interference spectral data undergoes axial pixel merging and lateral pixel merging processes respectively to generate low-resolution interference spectral data with reduced spatial resolution. The axial pixel merging process can be achieved through bandwidth narrowing; the lateral pixel merging process directly merges image pixels, such as 2-in-1 or 4-in-1.

[0024] S3 Low-resolution phase difference calculation: Based on the merged low-resolution data, repeat the steps. S 1 and S 2. Calculate the phase difference diagram after axial merging. Δφ a and its axially merged low-resolution phase quality distribution map D a And the phase difference map after horizontal merging Δ φl and its low-resolution phase quality distribution map after horizontal merging D l .

[0025] S4 Adaptive fusion: The optimal phase difference result is adaptively selected according to the following formula to generate the final output phase difference map: ; In the above formula, Do . T It can be considered as a judgment factor; Do . T = Do – Thr ; Thr For example, 3%.

[0026] The judgment logic of the judgment factor is: First, check the high-resolution results. Δφ _ o Whether the value of the phase quality distribution map is good enough, i.e., below the threshold. Thr If so, high-resolution results should be used first to preserve the best spatial resolution.

[0027] If the phase quality map value corresponding to the high-resolution result is poor, i.e., higher than the threshold, it indicates that there is severe sub-pixel decorrelation noise in that region. In this case, compare the phase quality map values ​​of the lateral and axial merged results, and select the lower-resolution phase difference result with better phase quality, i.e., lower phase quality map value, as the output for that region.

[0028] The aforementioned adaptive fusion algorithm compares phase results and their quality evaluation indicators, such as phase quality distribution maps, at different resolutions in real time, namely the original resolution, axial merging, and lateral merging, thereby dynamically and at the pixel level selecting the optimal phase data and achieving the best trade-off between signal-to-noise ratio and spatial resolution.

[0029] S5 Output: The final output is a fused phase difference map, which uses low-resolution data after pixel merging in areas with severe sub-pixel decorrelation, while retaining the original high-resolution data in other areas.

[0030] The method described in this embodiment directly reduces the cross-pixel speckle ratio that causes noise by pixel merging. R The method reduces the magnitude of decorrelation noise at its source, effectively suppressing sub-pixel decorrelation noise. Experimental data show that this method can further reduce phase noise in the boundary region by up to 19% on the basis of pixel-level compensation.

[0031] Furthermore, this invention does not simply perform global pixel merging. Instead, it uses an adaptive algorithm to enable low-resolution, high-SNR data only in regions with degraded SNR (regions corresponding to values ​​in the high phase quality distribution map), while retaining the original high-resolution data in regions with good SNR. This significantly improves overall phase quality while minimizing unnecessary resolution loss, adaptively balancing resolution and SNR.

[0032] Furthermore, this invention can serve as a post-processing module and can be seamlessly integrated into existing systems. PhS - OCT In the data processing workflow, it works in conjunction with various existing pixel-level decorrelation compensation algorithms, enhancing the performance of existing technologies and broadening their application scope. PhS - OCT Applicability in scenarios involving large deformation and high strain gradient. Specific Implementation Example 2: The present invention provides another embodiment. A phase-sensitive OCT The processing system employs the adaptive sub-pixel decorrelation method for optical coherence tomography as described in Specific Embodiment 1. Specific Implementation Example 3: The present invention also provides an embodiment: An electronic device includes: a storage medium and a processing unit; wherein the storage medium is used to store a computer program, and the processing unit exchanges data with the storage medium for executing the computer program during phase-sensitive optical coherence tomography to perform the steps of the method described in Specific Embodiment 1.

[0035] A computer-readable storage medium storing a computer program; when the computer program is run, it performs the steps of the method described in Specific Embodiment 1.

[0036] In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can transmit, propagate, or transfer a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wireline, optical fiber, etc. RF And so on, or any suitable combination of the above.

[0037] The above descriptions only cover a few specific embodiments of the present invention. However, the present invention is not limited thereto, and any variations that can be conceived by those skilled in the art should fall within the protection scope of the present invention. The above-mentioned serial numbers are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

Claims

1. An adaptive sub-pixel decorrelation method for optical coherence tomography, characterized in that, include: Obtain interference spectral data of the sample before and after deformation; High-resolution phase difference and high-resolution phase quality distribution map are obtained using the interference spectral data; and the interference spectral data are merged to obtain low-resolution phase difference and low-resolution phase quality distribution map after axial merging, as well as low-resolution phase difference and low-resolution phase quality distribution map after lateral merging. Based on the high-resolution phase difference and high-resolution phase quality distribution map, the low-resolution phase difference after axial merging, the low-resolution phase quality distribution map after axial merging, the low-resolution phase difference after lateral merging, and the low-resolution phase quality distribution map after lateral merging, the optimal phase difference result is adaptively selected to obtain the final output phase difference. Output the final phase difference diagram.

2. The adaptive sub-pixel decorrelation method for optical coherence tomography according to claim 1, characterized in that, The process of acquiring high-resolution phase difference and high-resolution phase quality distribution maps from the interferometric spectral data includes: The complex values ​​of the sample before and after deformation are obtained based on the interference spectral data. OCT The signal was analyzed to obtain a high-resolution phase map. The high-resolution phase difference and high-resolution phase quality distribution map corresponding to the high-resolution phase map are obtained by using a pixel-level decorrelation compensation method.

3. The adaptive sub-pixel decorrelation method for optical coherence tomography according to claim 1, characterized in that, The process of obtaining the low-resolution phase difference and the low-resolution phase quality distribution map after axial merging includes: The interferometric spectral data is used to perform axial pixel merging processing to obtain the corresponding axial low-resolution interferometric spectral data; The pixel-level decorrelation compensation method is used to process the axial low-resolution interferometric spectral data to obtain the axially merged low-resolution phase difference and the axially merged low-resolution phase quality distribution map.

4. The adaptive sub-pixel decorrelation method for optical coherence tomography according to claim 1, characterized in that, The process of obtaining the low-resolution phase difference after horizontal merging and the low-resolution phase quality distribution map after axial merging includes: The interferometric spectral data is used to perform lateral pixel merging processing to obtain the corresponding lateral low-resolution interferometric spectral data; The pixel-level decorrelation compensation method is used to process the lateral low-resolution interferometric spectral data to obtain the laterally merged low-resolution phase difference and the laterally merged low-resolution phase quality distribution map, respectively.

5. The adaptive sub-pixel decorrelation method for optical coherence tomography according to claim 1, characterized in that, The process of adaptively selecting the optimal phase difference result based on the high-resolution phase difference and high-resolution phase quality distribution map, the low-resolution phase difference after axial merging, the low-resolution phase quality distribution map after axial merging, and the low-resolution phase difference and low-resolution phase quality distribution map after lateral merging, to obtain the final output phase difference map includes: The final output phase difference map is generated by adaptively selecting the optimal phase difference result according to the following formula: ； in: Δφ o For high-resolution phase difference; D oT As a judgment factor; Δφ a This represents the low-resolution phase difference after axial merging. D a This is a low-resolution phase quality distribution map after axial merging; Δφ l This represents the low-resolution phase difference after horizontal merging. D l This is a low-resolution phase quality distribution map after horizontal merging.

6. The adaptive sub-pixel decorrelation method for optical coherence tomography according to claim 5, characterized in that: The judgment factor D oT for: D oT = Do – Thr ； in, D o A high-resolution phase quality distribution map; Thr The preset threshold; Based on the aforementioned judgment factors D oT Obtain the optimal spatial resolution.

7. The adaptive sub-pixel decorrelation method for optical coherence tomography according to claim 6, characterized in that, The judgment factor D oT To obtain the optimal spatial resolution, including: High-resolution phase quality distribution map corresponding to high-resolution phase difference D o With the threshold Thr The results are compared, and the high-resolution phase difference results are retained or the low-resolution phase difference results are used as the final output.

8. The adaptive sub-pixel decorrelation method for optical coherence tomography according to claim 7, characterized in that, The high-resolution phase quality distribution map corresponding to the high-resolution phase difference D o With the threshold Thr The comparison is performed, retaining the high-resolution result corresponding to the high-resolution phase difference or using the low-resolution phase difference result as the final output. include: High-resolution phase quality distribution map corresponding to high-resolution phase difference D o With the threshold Thr Comparison: If the high-resolution phase difference corresponds to the high-resolution phase quality distribution map D o Less than the threshold Thr If so, the corresponding high-resolution phase difference is selected as the output for that region; If the high-resolution phase difference corresponds to the high-resolution phase quality distribution map D o Greater than the threshold Thr If the low-resolution phase difference result is selected as the output for that region, then the output for that region will be chosen.

9. The adaptive sub-pixel decorrelation method for optical coherence tomography according to claim 8, characterized in that, The selection of low-resolution phase difference results as the output for this region includes: The low-resolution phase quality distribution maps after axial merging and the low-resolution phase quality distribution maps after lateral merging are obtained by using the low-resolution phase difference after axial merging or the low-resolution phase difference after lateral merging, respectively. By comparing the low-resolution phase quality distribution map after axial merging and the low-resolution phase quality distribution map after lateral merging, the low-resolution phase difference after axial merging or the low-resolution phase difference after lateral merging with better phase quality is taken as the output of this region.

10. A phase-sensitive OCT The processing system is characterized by: The adaptive sub-pixel decorrelation method for optical coherence tomography as described in any one of claims 1-9 is applied.

Images