Harmonic signal reconstruction method based on amplitude-phase joint filtering and product

Abstract

The application relates to the technical field of nonlinear optics, and discloses a second-harmonic signal reconstruction method and product based on amplitude-phase joint filtering, which comprises the following steps: performing second-harmonic polarization imaging on a sample by using a microscopic imaging system to obtain a plurality of second-harmonic intensity images; constructing a three-dimensional image stack based on the plurality of second-harmonic intensity images; performing Fourier transform on the three-dimensional image stack along a third dimension to obtain a target frequency domain data set; performing joint filtering processing on the amplitude and the phase in the target frequency domain data set, and reconstructing to obtain a plurality of initial time domain second-harmonic intensity images after denoising; and performing adaptive threshold denoising processing on the plurality of initial time domain second-harmonic intensity images to obtain a plurality of target time domain second-harmonic intensity images of the sample. Therefore, by implementing the application, the amplitude and the phase information are jointly processed, the target signal can be more accurately restored, and the imaging quality is improved.

Description

Technical Field

[0001] This invention relates to the field of nonlinear optics technology, specifically to a method and product for reconstructing second harmonic signals based on amplitude and phase joint filtering. Background Technology

[0002] In polarization second harmonic microscopy, the incident laser, after being shaped and polarized by an optical system, illuminates the sample. The nonlinear structures within the sample (such as collagen fibers) generate second harmonic signals, thus enabling imaging of its microstructure and spatial orientation. In traditional second harmonic polarization microscopy, the laser polarization angle is controlled by rotating a half-wave plate (λ / 2), thereby adjusting the generated second harmonic signal. By fitting a series of second harmonic intensity data at different polarization angles, the corresponding second harmonic polarization orientation angle can be extracted. However, image quality is often limited by noise during the imaging process, especially under low signal-to-noise ratio conditions. Noise not only affects image clarity but also interferes with subsequent analysis and diagnosis.

[0003] Traditional denoising methods (such as Gaussian filtering and bilateral filtering) can remove some noise, but they are not very effective at separating periodic signals (such as second harmonic signals) and random noise. This can lead to the loss of detailed information of the target signal during the denoising process, affecting the accuracy of imaging.

[0004] Therefore, how to preserve the periodicity and phase information of polarization second harmonic signals while removing noise has become a key problem that urgently needs to be solved in this field.

[0005] In the existing technology, Fourier transform-based denoising methods have been applied to some extent through frequency domain analysis, but they usually rely on fixed intensity thresholds or simple frequency filters, lacking adaptability and accuracy, and are particularly difficult to achieve effective noise suppression in complex backgrounds. Summary of the Invention

[0006] This invention provides a second harmonic signal reconstruction method and product based on amplitude and phase joint filtering, which solves the problems of poor separation effect of traditional denoising methods on periodic signals (such as second harmonic signals) and random noise, which affects the accuracy of imaging, and the lack of adaptability and accuracy of Fourier transform-based denoising methods, especially the difficulty in achieving effective noise suppression in complex backgrounds.

[0007] In a first aspect, the present invention provides a method for reconstructing a second harmonic signal based on joint amplitude and phase filtering, the method comprising:

[0008] Second harmonic polarization imaging of the sample was performed using a microscopic imaging system to obtain multiple second harmonic intensity images, each with a different illumination polarization angle. A three-dimensional image stack was constructed based on these multiple second harmonic intensity images. The three-dimensional image stack was then subjected to a Fourier transform along its third dimension to obtain a target frequency domain dataset. Joint filtering of the amplitude and phase values ​​in the target frequency domain dataset was performed, and multiple denoised initial time-domain second harmonic intensity images were reconstructed. Adaptive threshold denoising was then applied to these initial time-domain second harmonic intensity images to obtain multiple target time-domain second harmonic intensity images of the sample.

[0009] The second harmonic signal reconstruction method based on amplitude and phase joint filtering provided by this invention, through second harmonic polarization imaging of the sample using a microscopic imaging system, can ensure the integrity of the signal's periodic characteristics and avoid the loss of structural information caused by single-angle imaging. Furthermore, by constructing a three-dimensional image stack, it solves the problem that traditional two-dimensional data cannot simultaneously carry the spatial and polarization dimension correlation. Furthermore, through Fourier transform, the signal can be converted to the frequency domain, concentrating the second harmonic periodic signal at the fundamental frequency and dispersing noise across the entire frequency range, achieving preliminary separation of signal and noise. Furthermore, through joint filtering, both amplitude and phase characteristics can be preserved simultaneously, thus helping to effectively suppress noise in low signal-to-noise ratio environments, restoring the periodicity and spatial integrity of the target signal, and improving the signal-to-noise ratio and structural clarity of the initial time-domain image. Furthermore, through adaptive threshold denoising processing, residual background noise can be accurately removed while avoiding accidental deletion of the target signal, further enhancing image quality and recognizability. Therefore, by implementing this invention, through joint processing of amplitude and phase information, the target signal can be reconstructed more accurately, improving imaging quality.

[0010] In one alternative implementation, the three-dimensional image stack is Fourier transformed along the third dimension to obtain the target frequency domain dataset, including:

[0011] The three-dimensional image stack is subjected to Fourier transform along the third dimension to obtain the initial frequency domain dataset; the fundamental frequency component is extracted from the initial frequency domain dataset to obtain the target frequency domain dataset.

[0012] The second harmonic signal reconstruction method based on amplitude and phase joint filtering provided by this invention transforms the signal changes in the polarization angle dimension into quantifiable frequency components through Fourier transform, making the periodic characteristics of the signal explicit. Furthermore, by extracting the fundamental frequency component, it is possible to eliminate scattered noise components in the frequency domain and focus on the core frequency of the second harmonic signal, thus solving the problem of inaccurate noise suppression due to frequency domain filtering relying on a fixed threshold.

[0013] In one optional implementation, the fundamental frequency component is extracted from the initial frequency domain dataset to obtain the target frequency domain dataset, including:

[0014] Obtain the polarization angle modulation period value; based on the polarization angle modulation period value and frequency domain conjugate symmetry, extract the fundamental frequency component from the initial frequency domain dataset to obtain the target frequency domain dataset.

[0015] The second harmonic signal reconstruction method based on amplitude and phase joint filtering provided by this invention can avoid signal extraction deviations caused by misjudgment of the fundamental frequency position by obtaining the polarization angle modulation period value. Furthermore, by using the period value to lock the fundamental frequency position and combining conjugate symmetry to ensure the physical validity of the frequency domain signal, the extracted fundamental frequency component is ensured to be complete and distortion-free.

[0016] In one optional implementation, the amplitude and phase of the target frequency domain dataset are jointly filtered, and multiple denoised initial time-domain second harmonic intensity images are reconstructed, including:

[0017] The amplitude and phase of the target frequency domain data are jointly filtered to obtain multiple frequency domain signals; the multiple frequency domain signals are subjected to inverse Fourier transform and time domain reconstruction to obtain multiple time domain signals; based on the multiple time domain signals, multiple initial time domain second harmonic intensity images after denoising are determined.

[0018] The second harmonic signal reconstruction method based on amplitude and phase joint filtering provided by this invention overcomes the destruction of amplitude and phase coupling caused by traditional single filtering by performing joint filtering on the amplitude and phase of the target frequency domain data. It also suppresses high-frequency amplitude noise and phase entanglement noise, preserving the signal's intensity characteristics and spatial structure correlation, making it particularly suitable for extremely weak signal scenarios and improving the quality of the frequency domain signal. Furthermore, through inverse Fourier transform and time-domain reconstruction, the purified signal in the frequency domain can be restored into directly imageable time-domain intensity data, ensuring that the time-domain signal is free of imaginary part noise and that the intensity distribution is consistent with the actual structure of the sample. Furthermore, by transforming the abstract time-domain signal into an intuitive image form, the microstructural details of the second harmonic signal are preserved, significantly reducing the initial image noise compared to the original image.

[0019] In one optional implementation, adaptive threshold denoising is performed on multiple initial time-domain second harmonic intensity images to obtain multiple target time-domain second harmonic intensity images of the sample, including:

[0020] Gray-level histogram analysis was performed on multiple initial time-domain second harmonic intensity images to determine an adaptive threshold. Based on the adaptive threshold, the multiple initial time-domain second harmonic intensity images were denoised to obtain multiple target time-domain second harmonic intensity images of the sample.

[0021] The second harmonic signal reconstruction method based on amplitude and phase joint filtering provided by this invention dynamically determines the threshold based on the actual gray-level distribution of the image, avoiding the problems of denoising or incomplete denoising caused by traditional fixed thresholds, thus helping to accurately identify the intensity boundary between noise and signal. Furthermore, by combining adaptive thresholding for denoising processing, residual background noise in the initial image can be removed while completely preserving the target signal, improving the signal-to-noise ratio and contrast of the image, and making the microstructure of the sample clearer.

[0022] In an optional implementation, the method further includes: normalizing and contrast-enhancing multiple target time-domain second harmonic intensity images.

[0023] The second harmonic signal reconstruction method based on amplitude and phase joint filtering provided by this invention can unify the range of image pixel values ​​through normalization processing, eliminate the interference of pixel value magnitude differences on subsequent analysis, and improve data comparability. Furthermore, the contrast enhancement processing strengthens the grayscale difference between the target signal and the background, improves the image's recognizability, and makes tiny microstructures easier to observe.

[0024] In an alternative implementation, the method further includes: determining the polarization orientation of the sample based on multiple target time-domain second harmonic intensity images.

[0025] The second harmonic signal reconstruction method based on amplitude and phase joint filtering provided by this invention can accurately fit the polarization orientation angle of the sample through the high-quality image after denoising, which helps to realize the quantitative analysis of the microstructural features of the sample.

[0026] Secondly, the present invention provides a second harmonic signal reconstruction device based on amplitude and phase joint filtering, the device comprising:

[0027] The imaging module is used to perform second harmonic polarization imaging of the sample using a microscopic imaging system, obtaining multiple second harmonic intensity images, where the illumination polarization angle corresponding to each second harmonic intensity image is different; the construction module is used to construct a three-dimensional image stack based on multiple second harmonic intensity images; the transformation module is used to perform Fourier transform on the three-dimensional image stack along the third dimension to obtain the target frequency domain dataset; the filtering module is used to perform joint filtering on the amplitude and phase in the target frequency domain dataset and reconstruct multiple denoised initial time-domain second harmonic intensity images; the processing module is used to perform adaptive threshold denoising on the multiple initial time-domain second harmonic intensity images to obtain multiple target time-domain second harmonic intensity images of the sample.

[0028] Thirdly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the second harmonic signal reconstruction method based on amplitude-phase joint filtering according to the first aspect or any corresponding embodiment described above.

[0029] Fourthly, the present invention provides a computer program product, including computer instructions for causing a computer to execute the second harmonic signal reconstruction method based on amplitude and phase joint filtering as described in the first aspect or any corresponding embodiment. Attached Figure Description

[0030] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0031] Figure 1 This is a schematic diagram of an application scenario according to an embodiment of the present invention;

[0032] Figure 2 This is a schematic diagram of the first process of a second harmonic signal reconstruction method based on amplitude and phase joint filtering according to an embodiment of the present invention;

[0033] Figure 3 This is a schematic diagram of the second process of the second harmonic signal reconstruction method based on amplitude and phase joint filtering according to an embodiment of the present invention;

[0034] Figure 4 This is a schematic diagram of the third process of the second harmonic signal reconstruction method based on amplitude and phase joint filtering according to an embodiment of the present invention;

[0035] Figure 5 This is a flowchart illustrating the second harmonic signal reconstruction method based on amplitude-phase joint filtering according to an embodiment of the present invention.

[0036] Figure 6 This is a schematic diagram illustrating the process of reconstructing a second harmonic signal based on amplitude-phase joint filtering of several original data images corresponding to each polarization angle according to an embodiment of the present invention.

[0037] Figure 7A This is a schematic diagram of the structural strength indicated by the arrow and the strength at the box in m original images according to an embodiment of the present invention;

[0038] Figure 7BThis is a schematic diagram of the structural strength indicated by the arrow and the strength at the box in n enhanced images according to an embodiment of the present invention;

[0039] Figure 8 This is a schematic diagram illustrating the effect of the second harmonic signal reconstruction method based on amplitude-phase joint filtering according to an embodiment of the present invention;

[0040] Figure 9 This is a structural block diagram of a second harmonic signal reconstruction device based on amplitude and phase joint filtering according to an embodiment of the present invention;

[0041] Figure 10 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.

[0044] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0045] As an optional application scenario of this invention, the specific application environment architecture or specific hardware architecture on which the second harmonic signal reconstruction method based on amplitude and phase joint filtering depends is described here. Figure 1 As shown, the architecture system may include at least one terminal device and at least one server. Figure 1 The system is illustrated in the example, which includes a computer 101, a mobile terminal 102, and a server 103, and the terminal devices such as the computer 101 and the mobile terminal 102 are connected to the server 103 through a network 110.

[0046] Specifically, the terminal device can be a smartphone, tablet, laptop, PDA, desktop computer, game console, smart TV, smart wearable device, in-vehicle terminal, VR (Virtual Reality) device, AR (Augmented Reality) device, etc. Server 103 can be a standalone physical server, a server cluster, a distributed system, or a cloud server providing cloud services. Network 110 can be a wired or wireless network, examples of which include, but are not limited to, the Internet, corporate intranet, local area network, wide area network, mobile communication network, and combinations thereof.

[0047] This invention provides a method for reconstructing second harmonic signals based on amplitude and phase joint filtering. By performing joint filtering on amplitude and phase, the method can accurately restore the target signal and improve imaging quality.

[0048] According to an embodiment of the present invention, a method for reconstructing a second harmonic signal based on amplitude and phase joint filtering is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0049] This embodiment provides a second harmonic signal reconstruction method based on amplitude and phase joint filtering, which can be used in the aforementioned mobile terminals, such as mobile phones and tablets. Figure 2 This is a flowchart of a second harmonic signal reconstruction method based on amplitude and phase joint filtering according to an embodiment of the present invention, as shown below. Figure 2 As shown, the process includes the following steps:

[0050] Step S201: Use a microscopic imaging system to perform second harmonic polarization imaging on the sample to obtain multiple second harmonic intensity images.

[0051] In one optional embodiment, the microscopic imaging system refers to a comprehensive device that combines optical microscopy with digital imaging technology to achieve multi-scale observation and image recording of samples. In this embodiment, it is a second harmonic generation (SHG) microscopic imaging system, which is a hardware combination for realizing the excitation, detection, and imaging of second harmonic signals.

[0052] In one optional embodiment, the second harmonic intensity image represents a visualized grayscale image formed by the distribution of second harmonic signal intensity of a sample under laser excitation at a specific polarization angle. The grayscale value of each pixel in the image directly corresponds to the intensity of the second harmonic signal generated at that location on the sample; that is, the stronger the signal, the higher the grayscale value.

[0053] Furthermore, the illumination polarization angle corresponding to each second harmonic intensity image is different.

[0054] In one optional embodiment, a second harmonic microscopy system with adjustable polarization angle is used to image the sample multiple times under different illumination polarization angles, thereby obtaining multiple corresponding second harmonic intensity images.

[0055] In one optional embodiment, a half-wave plate is added to the laser optical path of a conventional second harmonic microscopy imaging system. By rotating the half-wave plate at different angles, the laser polarization direction is adjusted, and the sample is excited to generate a second harmonic signal under each polarization direction. The corresponding intensity image is then acquired, thereby obtaining multiple corresponding second harmonic intensity images. By adjusting the angle of the half-wave plate, second harmonic signals can be obtained in different directions, thus enhancing the flexibility and accuracy of the imaging system.

[0056] For example, a half-wave plate is inserted between the laser source and the microscope objective in a traditional second harmonic microscopy system. It is necessary to ensure that the optical axis of the half-wave plate is coaxial with the laser beam path to avoid polarization adjustment errors caused by optical axis misalignment.

[0057] Furthermore, the rotation angle range and angle interval of the half-wave plate are set to generate an angle sequence.

[0058] Furthermore, for each target angle in the angle sequence, the following operations are performed sequentially:

[0059] (1) Polarization direction adjustment: The half-wave plate is rotated to the current target angle by controlling the precision motor, and then left to stand still for 1-2 seconds to stabilize the optical path and ensure that the laser polarization direction is consistent with the target angle;

[0060] (2) Excitation of second harmonic signal: After the laser is polarized by a half-wave plate, it is focused onto the sample through a microscope objective. The ordered structure in the sample is excited to generate a second harmonic signal.

[0061] (3) Signal acquisition and imaging: After the second harmonic signal is collected by the objective lens, it enters the detector through the filter. The detector converts the optical signal into an electrical signal and finally generates a second harmonic intensity image corresponding to the current polarization angle.

[0062] Step S202: Construct a three-dimensional image stack based on multiple second harmonic intensity images.

[0063] In one optional embodiment, the three-dimensional image stack represents a three-dimensional data matrix formed by integrating multiple second harmonic intensity images according to the spatial pixel-polarization angle dimension, wherein the dimension is defined as follows: the X-axis and Y-axis are the image pixel coordinates, and the Z-axis is the number of polarization angle modulations.

[0064] In one optional embodiment, the second harmonic intensity images corresponding to different polarization angles are integrated into a unified three-dimensional data matrix, i.e., a three-dimensional image stack, according to the correspondence between spatial pixels and polarization angles.

[0065] Step S203: Perform a Fourier transform on the three-dimensional image stack along the third dimension to obtain the target frequency domain dataset.

[0066] In an alternative embodiment, Fourier transform refers to one-dimensional fast Fourier transform (1D-FFT), which is a mathematical operation that converts the time-domain signal (polarization angle dimension) of the third dimension in a three-dimensional image stack into a frequency-domain signal.

[0067] In one optional embodiment, the target frequency domain dataset represents the frequency domain data matrix obtained after Fourier transform and fundamental frequency component extraction of the 3D image stack. Only the fundamental frequency and its conjugate component corresponding to the periodicity of the second harmonic signal are retained, while other frequency components (noise) are set to zero. It may contain the amplitude and phase information of the signal and satisfies frequency domain conjugate symmetry.

[0068] In one alternative embodiment, a Fourier transform is performed on the third dimension of the three-dimensional image stack, and the changes in second harmonic intensity under different polarization angles are converted into frequency domain distributions. Then, by extracting the fundamental frequency component, a frequency domain dataset containing only the target signal energy, i.e., the target frequency domain dataset, can be obtained.

[0069] Step S204: Perform joint filtering on the amplitude and phase of the target frequency domain dataset, and reconstruct multiple initial time-domain second harmonic intensity images after denoising.

[0070] In one optional embodiment, a collaborative processing filtering strategy is used to process the amplitude and phase information in the target frequency domain dataset. This suppresses noise while preserving the coupling relationship between the two. Then, the purified frequency domain signal is converted back to the time domain through inverse transformation, resulting in multiple initial time domain second harmonic intensity images with significantly reduced noise.

[0071] In one optional embodiment, a combined time-domain and frequency-domain filtering method is used. The image signal is initially smoothed in the time domain (such as median filtering, mean filtering, etc.), and residual noise is removed in the frequency domain. More refined frequency-domain filtering (such as bandpass filtering, window function, etc.) is used to suppress noise.

[0072] Furthermore, image quality can be further optimized by selectively enhancing signals in specific frequency bands. Specifically, enhancing the signal in a specific frequency band can improve detail by boosting the high-frequency components based on the Fourier transform results, while weakening the low-frequency components to reduce noise.

[0073] Step S205: Perform adaptive threshold denoising on multiple initial time-domain second harmonic intensity images to obtain multiple target time-domain second harmonic intensity images of the sample.

[0074] In one alternative embodiment, since the signal strength and noise strength are similar under initial conditions, traditional fixed threshold denoising methods may not be effective in removing noise. Therefore, in this embodiment, an adaptive threshold can be used to remove residual background noise from the image, ultimately obtaining a target time-domain second harmonic intensity image that retains only the effective second harmonic signal.

[0075] In one optional embodiment, the difference between small grayscale fluctuations in the signal region and large grayscale fluctuations in the noise region can be utilized to dynamically generate the denoising threshold of each pixel based on the standard deviation of the local region surrounding each pixel, thereby avoiding misjudgment of local weak signals by the global threshold.

[0076] In an optional embodiment, the difference between signal pixels and neighboring pixels having strong grayscale correlation and noise pixels having weak correlation with their neighbors can be utilized to generate a threshold by calculating the similarity between pixels and their neighbors, thereby accurately distinguishing between signals and noise.

[0077] The second harmonic signal reconstruction method based on amplitude and phase joint filtering provided in this embodiment, through second harmonic polarization imaging of the sample using a microscopic imaging system, can ensure the integrity of the signal's periodic characteristics and avoid the loss of structural information caused by single-angle imaging. Furthermore, by constructing a three-dimensional image stack, it solves the problem that traditional two-dimensional data cannot simultaneously carry the spatial and polarization dimension correlation. Furthermore, through Fourier transform, the signal can be converted to the frequency domain, concentrating the second harmonic periodic signal at the fundamental frequency and dispersing noise across the entire frequency range, achieving preliminary separation of signal and noise. Furthermore, joint filtering can simultaneously preserve amplitude and phase characteristics, thus helping to effectively suppress noise in low signal-to-noise ratio environments, restoring the periodicity and spatial integrity of the target signal, and improving the signal-to-noise ratio and structural clarity of the initial time-domain image. Furthermore, through adaptive threshold denoising processing, residual background noise can be accurately removed while avoiding accidental deletion of the target signal, further enhancing image quality and recognizability. Therefore, by implementing this invention, through joint processing of amplitude and phase information, the target signal can be more accurately reconstructed, improving imaging quality.

[0078] This embodiment provides a second harmonic signal reconstruction method based on amplitude and phase joint filtering, which can be used in the aforementioned mobile terminals, such as mobile phones and tablets. Figure 3 This is a flowchart of a second harmonic signal reconstruction method based on amplitude and phase joint filtering according to an embodiment of the present invention, as shown below. Figure 3 As shown, the process includes the following steps:

[0079] Step S301: Second harmonic polarization imaging of the sample is performed using a microscopic imaging system to obtain multiple second harmonic intensity images. For details, please refer to [link to relevant documentation]. Figure 2Step S201 of the illustrated embodiment will not be described again here.

[0080] Step S302: Construct a three-dimensional image stack based on multiple second harmonic intensity images. See details below. Figure 2 Step S202 of the illustrated embodiment will not be described again here.

[0081] Step S303: Perform a Fourier transform on the three-dimensional image stack along the third dimension to obtain the target frequency domain dataset.

[0082] Specifically, step S303 includes:

[0083] Step S3031: Perform a Fourier transform on the three-dimensional image stack along the third dimension to obtain the initial frequency domain dataset.

[0084] In one optional embodiment, the initial frequency domain dataset may include a DC component (i.e., the mean intensity of all images), positive frequency components (from low to high frequencies), and negative frequency components. The negative frequency components are conjugate symmetrical with the positive frequency components.

[0085] In one optional embodiment, a one-dimensional transformation is performed on the Z-axis, or third dimension, of the three-dimensional image stack using Fourier transform. Each spatial pixel corresponds to a frequency domain signal sequence of length N.

[0086] Furthermore, the initial frequency domain dataset obtained after the transformation has the following characteristics:

[0087] (1) Frequency domain data size: [image height, image width, number of polarization angle modulations].

[0088] (2) Frequency component distribution: [image height, image width, 1]: DC component - representing the mean of all images; [image height, image width, 2 to N / 2]: positive frequency components (from low frequency to high frequency); [image height, image width, N / 2+1 to N]: negative frequency components (symmetric to positive frequency).

[0089] Furthermore, due to the symmetry of FFT, the negative frequency component is conjugate symmetric with the positive frequency component. When the dimension of the total polarization sequence is N (the total length of the polarization angle modulation sequence), the index k and the index N+2-k are conjugates.

[0090] Step S3032: Extract the fundamental frequency component from the initial frequency domain dataset to obtain the target frequency domain dataset.

[0091] In one optional embodiment, by filtering out the fundamental frequency and its conjugate component representing the target signal from the initial frequency domain data, other frequency components corresponding to noise can be eliminated, thereby obtaining a target frequency domain dataset that retains only the effective signal, thus achieving frequency domain separation of signal and noise.

[0092] In some optional implementations, step S3032 above includes:

[0093] Step a1: Obtain the polarization angle modulation period value.

[0094] Step a2: Based on the polarization angle modulation period value and frequency domain conjugate symmetry, the fundamental frequency component is extracted from the initial frequency domain dataset to obtain the target frequency domain dataset.

[0095] In an optional embodiment, the polarization angle modulation period value represents the number of angle intervals during which the intensity of the second harmonic signal changes periodically during the polarization angle adjustment process. That is, for every T (period value) polarization angles adjusted, the intensity of the second harmonic generated by the sample will repeat the fluctuation pattern once.

[0096] In an optional embodiment, frequency domain conjugate symmetry means that after the real-valued time-domain signal is Fourier transformed, its frequency domain signal satisfies the complex conjugate symmetry relationship.

[0097] In an alternative embodiment, the number of periods P of polarization angle modulation can be obtained based on known characteristics of the sample or preliminary experiments.

[0098] Furthermore, based on the polarization angle modulation period value, the fundamental frequency index in the frequency domain can be determined as P, and then, combined with conjugate symmetry, its conjugate component index can be determined as N+2-P.

[0099] Furthermore, a new frequency domain matrix of the same size as the initial frequency domain data is constructed and initialized, and the original frequency domain signal is filled only at the fundamental frequency index P and the conjugate component index N+2-P, while all other frequency components are set to zero.

[0100] Furthermore, the frequency domain conjugate symmetry is verified to ensure that the fundamental frequency and the conjugate components satisfy the conjugate relationship of equal real parts and opposite imaginary parts. This ensures that the subsequent inverse transformation yields a real-valued time-domain signal, and the final constructed new frequency domain matrix is ​​the target frequency domain dataset.

[0101] Step S304: Perform joint filtering on the amplitude and phase of the target frequency domain dataset, and reconstruct multiple initial time-domain second harmonic intensity images after denoising.

[0102] Specifically, step S304 includes:

[0103] Step S3041: Perform joint filtering on the amplitude and phase of the target frequency domain data to obtain multiple frequency domain signals.

[0104] In one optional embodiment, Gaussian filtering is applied to the amplitude and phase of the target frequency domain data, while smoothing the real and imaginary parts of the frequency domain signal. This process suppresses noise while maintaining the coupling relationship between amplitude and phase, ultimately yielding a denoised frequency domain signal.

[0105] For example, the amplitude and phase of each spatial pixel are extracted from the target frequency domain data.

[0106] Furthermore, a Gaussian filter (standard deviation σ1 ≥ 0.5) is applied to the amplitude matrix to remove high-frequency random noise while preserving the main signal structure. A smaller standard deviation σ1 can be used to minimize the loss of detail.

[0107] Simultaneously, a Gaussian filter is applied to the phase matrix. A larger standard deviation can be selected to remove low-frequency phase noise and preserve the overall image structure.

[0108] Furthermore, Gaussian filtering is applied to the real and imaginary parts of the frequency domain signal respectively to maintain the coupling relationship between amplitude and phase and avoid information loss caused by individual filtering.

[0109] Finally, the filtered amplitude and phase are recombine into a complex form of frequency domain signal, i.e., multiple frequency domain signals.

[0110] Step S3042: Perform inverse Fourier transform and time-domain reconstruction on multiple frequency domain signals to obtain multiple time-domain signals.

[0111] In one optional embodiment, an inverse Fourier transform is performed on the filtered frequency domain signal along the polarization angle dimension to convert the purified signal in the frequency domain back into a polarization angle-intensity time domain sequence, thereby obtaining a denoised time domain signal.

[0112] For example, the obtained frequency domain signal is subjected to a one-dimensional inverse Fourier transform along the polarization angle dimension, i.e., the third dimension, using the inverse Fourier transform. Furthermore, since the frequency domain maintains conjugate symmetry, the result obtained after the inverse transform is a real-valued signal, and each spatial pixel corresponds to a time-domain sequence of polarization angle-intensity.

[0113] Furthermore, the temporal sequences of all spatial pixels are integrated into a three-dimensional data matrix, i.e., multiple temporal signals, and this dimension is consistent with the original image stack ([image height, image width, number of polarization angle modulations]).

[0114] Step S3043: Determine multiple initial time-domain second harmonic intensity images after denoising based on multiple time-domain signals.

[0115] In one alternative embodiment, multiple initial time-domain second harmonic intensity images can be obtained by directly mapping the time-domain signal to a visualized second harmonic intensity image.

[0116] For example, in the third dimension of the time-domain signal, the intensity values ​​of all spatial pixels under that index are extracted to form a two-dimensional matrix. Each index corresponds to a polarization angle.

[0117] Furthermore, each two-dimensional matrix corresponds to a second harmonic intensity image at a specific polarization angle. Therefore, by arranging all the two-dimensional matrices corresponding to the polarization angles in order, multiple initial time-domain second harmonic intensity images are obtained after denoising.

[0118] Step S305 involves performing adaptive threshold denoising on multiple initial time-domain second harmonic intensity images to obtain multiple target time-domain second harmonic intensity images of the sample. For details, please refer to [link to relevant documentation]. Figure 2 Step S205 of the illustrated embodiment will not be described again here.

[0119] The second harmonic signal reconstruction method based on amplitude and phase joint filtering provided in this embodiment transforms the signal changes in the polarization angle dimension into quantifiable frequency components through Fourier transform, making the periodic characteristics of the signal explicit. Furthermore, by obtaining the polarization angle modulation period value, signal extraction deviations caused by misjudgment of the fundamental frequency position can be avoided. Furthermore, by using the period value to lock the fundamental frequency position and combining conjugate symmetry to ensure the physical validity of the frequency domain signal, the extracted fundamental frequency component is ensured to be complete and distortion-free. Furthermore, by performing joint filtering on the amplitude and phase of the target frequency domain data, the destruction of amplitude and phase coupling caused by traditional single filtering is overcome, while suppressing high-frequency amplitude noise and phase entanglement noise, preserving the signal's intensity characteristics and spatial structure correlation, making it particularly suitable for extremely weak signal scenarios and improving the quality of the frequency domain signal. Furthermore, through inverse Fourier transform and time-domain reconstruction, the purified signal in the frequency domain can be restored to directly imageable time-domain intensity data, ensuring that the time-domain signal has no imaginary part noise and that the intensity distribution is consistent with the actual structure of the sample. Furthermore, the abstract time-domain signal is transformed into an intuitive image form, preserving the microstructural details of the second harmonic signal, and significantly reducing the noise of the initial image compared to the original image.

[0120] This embodiment provides a second harmonic signal reconstruction method based on amplitude and phase joint filtering, which can be used in the aforementioned mobile terminals, such as mobile phones and tablets. Figure 4 This is a flowchart of a second harmonic signal reconstruction method based on amplitude and phase joint filtering according to an embodiment of the present invention, as shown below. Figure 4 As shown, the process includes the following steps:

[0121] Step S401: Second harmonic polarization imaging of the sample is performed using a microscopic imaging system to obtain multiple second harmonic intensity images. For details, please refer to [link to relevant documentation]. Figure 2 Step S201 of the illustrated embodiment will not be described again here.

[0122] Step S402: Construct a three-dimensional image stack based on multiple second harmonic intensity images. See details below. Figure 2 Step S202 of the illustrated embodiment will not be described again here.

[0123] Step S403: Perform a Fourier transform on the 3D image stack along the third dimension to obtain the target frequency domain dataset. See details below. Figure 3 Step S303 of the illustrated embodiment will not be described again here.

[0124] Step S404 involves performing joint filtering on the amplitude and phase of the target frequency domain dataset, and reconstructing multiple initial time-domain second harmonic intensity images after denoising. For details, please refer to [link to relevant documentation]. Figure 3 Step S304 of the illustrated embodiment will not be described again here.

[0125] Step S405: Perform adaptive threshold denoising on multiple initial time-domain second harmonic intensity images to obtain multiple target time-domain second harmonic intensity images of the sample.

[0126] Specifically, step S405 includes:

[0127] Step S4051: Perform grayscale histogram analysis on multiple initial time-domain second harmonic intensity images and determine the adaptive threshold.

[0128] In one alternative embodiment, the histogram can reveal the gray-level distribution in the image, particularly the intensity difference between background noise and the target signal. Therefore, by analyzing the histogram of the initial time-domain second harmonic intensity image, the intensity range of the noise and signal can be determined. Further, the intensity value corresponding to the first distribution peak (noise distribution) is extracted from the histogram and used as an adaptive threshold.

[0129] For example, for each initial time-domain second harmonic intensity image, the grayscale value distribution of all its pixels is statistically analyzed to generate a corresponding grayscale histogram. The horizontal axis represents the grayscale value, and the vertical axis represents the number of pixels corresponding to that grayscale value.

[0130] Furthermore, the histogram is analyzed to extract the first distribution peak. This peak corresponds to the main intensity range of the background noise, since the number of noise pixels far exceeds the number of signal pixels at low signal-to-noise ratios.

[0131] Furthermore, the grayscale value corresponding to the peak of this distribution is used as an adaptive threshold. This adaptive threshold represents the boundary between signal and noise; pixels below this value are likely noise, while pixels above this value are the target signal.

[0132] Step S4052: Based on the adaptive threshold, denoise the multiple initial time-domain second harmonic intensity images to obtain multiple target time-domain second harmonic intensity images of the sample.

[0133] In one alternative embodiment, an adaptive threshold is used to filter and retain target signal pixels and remove noise pixels, ultimately obtaining a high-purity second harmonic intensity image.

[0134] For example, iterate through all pixels of each initial temporal image and determine its grayscale value.

[0135] Furthermore, if the pixel grayscale value is lower than the adaptive threshold, it is determined to be background noise and its grayscale value is set to 0; if the pixel grayscale value is higher than or equal to the adaptive threshold, it is determined to be the target signal and its original grayscale value is retained.

[0136] Furthermore, the filtered pixel matrix is ​​integrated into a new image and arranged in order of polarization angle to obtain multiple target time-domain second harmonic intensity images of the sample.

[0137] In some optional implementations, the method further includes: normalizing and contrast-enhancing multiple target time-domain second harmonic intensity images.

[0138] In one alternative embodiment, by mapping the pixel values ​​of the target time-domain image to a uniform numerical range, the problem of differences in pixel value magnitude between different images can be solved, thereby improving the image's visualization effect and detail clarity.

[0139] For example, iterate through a single target temporal image to obtain the minimum and maximum gray values ​​of all pixels. Then, for the gray value of each pixel in the image, map it to the target range, typically between [0, 255] or [0, 1].

[0140] In one alternative embodiment, by adjusting the grayscale distribution of pixels, the brightness difference between the signal area and the background can be increased, thus solving the problem of blurred details caused by low contrast.

[0141] In one alternative embodiment, an adaptive histogram equalization method is used to enhance the contrast of the image.

[0142] For example, the target temporal image is divided into multiple non-overlapping small regions. Then, a gray-level histogram is calculated separately for each sub-block, and the histogram is equalized.

[0143] Furthermore, the equalized sub-blocks are stitched together to form a complete image, and the boundaries between blocks are smoothed to avoid abrupt changes in grayscale between sub-blocks.

[0144] Furthermore, observe the processed image. If the clarity of signal details is improved, then contrast enhancement is complete; if the effect is insufficient, the sub-block size can be adjusted and the operation repeated.

[0145] In some alternative implementations, the method further includes determining the polarization orientation of the sample based on multiple target time-domain second harmonic intensity images.

[0146] In an optional embodiment, by utilizing the difference in second harmonic intensity of the target time-domain image at different polarization angles, the polarization orientation angle of the ordered structure in the sample can be fitted, thereby reflecting the microscopic arrangement direction of the sample.

[0147] For example, in the target time-domain image, a local area with clear signal is selected, avoiding areas with residual noise or weak signal. Then, for the pixels in that area, the second harmonic intensity values ​​at all polarization angles are statistically analyzed to obtain the variation curve between polarization angle and intensity (I).

[0148] Furthermore, the relationship between the second harmonic signal intensity and the polarization angle follows the cosine square law, thus fitting the curve between the polarization orientation angle of the sample and the maximum signal intensity.

[0149] Furthermore, by fitting, the polarization orientation angle of the corresponding sample, i.e., the polarization orientation of the sample in that region, can be obtained.

[0150] Furthermore, the above steps are repeated for multiple feature regions in the image, and then the average value of the polarization orientation angle is taken as the overall polarization orientation of the sample.

[0151] The second harmonic signal reconstruction method based on amplitude and phase joint filtering provided in this embodiment dynamically determines the threshold based on the actual gray-level distribution of the image, avoiding the problems of denoising or incomplete denoising caused by traditional fixed thresholds, thus helping to accurately identify the intensity boundary between noise and signal. Furthermore, by combining adaptive thresholding for denoising, residual background noise in the initial image can be removed while completely preserving the target signal, improving the signal-to-noise ratio and contrast of the image, making the sample's microstructure clearer. Furthermore, normalization processing unifies the range of image pixel values, eliminating the interference of pixel value magnitude differences on subsequent analysis and improving data comparability. Furthermore, contrast enhancement processing strengthens the gray-level difference between the target signal and the background, improving image recognizability and making minute microstructures easier to observe. Furthermore, the high-quality denoised image allows for accurate fitting of the sample's polarization orientation angle, facilitating the quantitative analysis of the sample's microstructural features.

[0152] In one instance, such as Figure 5As shown, a second harmonic signal reconstruction method based on amplitude-phase joint filtering is provided. Based on the amplitude-phase joint filtering strategy, it can effectively extract periodic signals from noise and preserve the detailed information (amplitude and phase information) of the image. It is particularly suitable for image reconstruction with low signal-to-noise ratio in optical imaging.

[0153] Furthermore, traditional filtering methods typically process amplitude or phase separately, neglecting their combined effect. This example proposes for the first time a joint filtering approach that combines amplitude and phase information, simultaneously smoothing both the amplitude and phase of the signal in the frequency domain. This joint filtering effectively removes high-frequency noise while enhancing the target signal and preserving its periodicity.

[0154] In an optional implementation, the core technical solution of this example includes the following steps:

[0155] 1. Laser optical path and half-wave plate adjustment:

[0156] In traditional second-harmonic generation (HHM) microscopy systems, a half-wave plate is added to the laser path. By rotating the half-wave plate at different angles, the laser polarization direction is adjusted, thereby generating second-harmonic signals at different angles. By adjusting the angle of the half-wave plate, second-harmonic signals can be obtained in different directions, thus enhancing the flexibility and accuracy of the imaging system.

[0157] 2. Construct the image spectrum stack:

[0158] In this step, a spectral stack of the image is constructed by performing a Fourier transform on the input image (an image sequence modulated by polarization angle). This spectral stack contains information about the image at various frequency components, helping to separate the different frequency domain characteristics of the signal and noise. First, the image sequence data is read, and an image stack is constructed (X and Y axes represent image pixel coordinates, and the Z axis represents the number of polarization angle modulations). The second harmonic signal is then subjected to a Fourier transform along the Z-axis (number of polarization angles) to obtain the spectral data. Different frequency components in the spectrum represent different details in the image, including signal and noise components. In the frequency domain, a joint filtering strategy is used for smoothing, combining the signal's amplitude and phase information. Compared to traditional single amplitude or phase filtering, joint filtering can better preserve periodic signals while removing noise. Its specific operation is as follows:

[0159] After performing a Fourier transform on the three-dimensional image stack along the third dimension (Z-axis / polarization sequence dimension), the resulting frequency domain data has the following characteristics:

[0160] (1) Frequency domain data size: [image height, image width, number of polarization angle modulations].

[0161] (2) Frequency component distribution: [image height, image width, 1]: DC component - representing the mean of all images; [image height, image width, 2 to N / 2]: positive frequency components (from low frequency to high frequency); [image height, image width, N / 2+1 to N]: negative frequency components (symmetric to positive frequency).

[0162] Due to the symmetry of FFT, the negative frequency components are conjugate symmetric with the positive frequency components. When the total polarization sequence dimension is N (the total length of the polarization angle modulation sequence), the index k and the index N+2-k are conjugates.

[0163] In the frequency domain, the modulated signal manifests as a specific frequency component (fundamental frequency), while noise signals are typically distributed across all frequencies, exhibiting dispersed energy. Separation of signal and noise is achieved by extracting the fundamental frequency component. The specific operation involves constructing a new frequency domain matrix, retaining only the fundamental frequency P (where P is the number of periods in the polarization angle modulation) and its conjugate component N+2-P (where the polarization angle modulation contains P periods), while setting other frequency components to zero. This is equivalent to applying bandpass filtering in the frequency domain. Furthermore, the frequency domain conjugate symmetry is maintained to ensure that a real-valued signal is obtained after the inverse transform.

[0164] 3. Amplitude-phase joint filtering:

[0165] Amplitude filtering: In the frequency domain, the amplitude of an image reflects the strength of the signal. Applying Gaussian filtering can remove high-frequency noise (such as random noise) while preserving the main signal structure, thus improving the signal-to-noise ratio of the image.

[0166] Phase filtering: The phase contains spatial structural information of an image. Phase noise can cause the geometric structure of the image to shift, affecting the reconstruction effect.

[0167] Gaussian filtering is applied separately to the amplitude and phase to smooth the signal and suppress noise. The standard deviation σ1 controls the degree of smoothing; a larger standard deviation results in a stronger smoothing effect and more significant noise removal, but may also lead to the loss of some details. The value of the Gaussian filter parameter σ1 is determined based on the structural characteristics of the actual imaging object; generally, σ1 is set no less than 0.5. In most cases, the noise levels and characteristics of amplitude and phase are similar. The same standard deviation helps ensure balanced noise reduction effects for amplitude and phase within the same image. Different standard deviations can also be set to optimize the filtering effects of amplitude and phase separately. Amplitude filtering can use a smaller standard deviation to preserve as much detail as possible, removing only noise. Phase filtering can use a larger standard deviation to remove low-frequency noise in the phase while maintaining the overall structure of the image.

[0168] Gaussian filtering is applied separately to the real and imaginary parts of the image to smooth spatial noise, maintain the coupling relationship between amplitude and phase, and avoid information loss associated with traditional methods (direct filtering). This method is particularly suitable for enhancing extremely weak signals (where the signal strength is roughly equivalent to the strength of random noise) (low signal-to-noise ratio images). Direct filtering typically refers to filtering the amplitude or phase of the image independently, or performing global processing on the image data directly in the frequency domain. For example, Gaussian filtering can be applied directly to the amplitude portion of the image, or spatial filtering can be applied to the entire image. This method focuses on noise removal, but may sometimes affect the structural information of the image.

[0169] 4. Reconstruction and Signal Enhancement:

[0170] After amplitude and phase filtering, the processed spectral signal is converted back to the time domain using an inverse Fourier transform, thus reconstructing the denoised image. The processed spectral signal is then reconstructed back to the time domain using an inverse Fourier transform to obtain the denoised second harmonic image. At this point, the amplitude of the target structure is almost identical to its amplitude before processing, while the amplitude of the noise region is significantly suppressed, resulting in a marked improvement in signal quality in the image. This image effectively preserves the detailed information of the periodic signal and removes background noise, ensuring signal accuracy.

[0171] 5. Adaptive threshold denoising:

[0172] Since the signal strength and noise strength are similar under initial conditions, traditional fixed-threshold denoising methods may not be effective in removing noise. Therefore, after amplitude-phase joint filtering, adaptive threshold denoising is introduced, which adaptively selects an appropriate threshold based on the distribution of the image.

[0173] First, a grayscale histogram analysis is performed on the reconstructed image (the denoised image). The histogram reveals the grayscale distribution in the image, especially the intensity difference between background noise and the target signal. By analyzing the image's histogram, the intensity range of noise and signal can be determined. The intensity value corresponding to the first distribution peak (noise distribution) is extracted from the histogram and used as an adaptive threshold. This threshold represents the boundary between signal and noise. Applying the adaptive threshold, pixels below the threshold (i.e., noise regions) are set to zero, while pixels above the threshold (i.e., signal regions) are retained.

[0174] This effectively removes background noise while preserving the valid signal in the image. Finally, by using the signal-to-noise ratio (SNR) as an evaluation metric, the amplitude-phase joint filtering and adaptive thresholding denoising method of this invention demonstrates significant effects in denoising and signal enhancement under low SNR conditions. The improved SNR demonstrates signal enhancement and noise suppression, enhancing image quality and recognizability.

[0175] 6. Post-processing steps: In order to further improve the image quality, after signal enhancement, post-processing is usually required.

[0176] Normalization: Normalizing the reconstructed image ensures that the pixel values ​​are within a suitable range, typically between [0, 255] or [0, 1]. Normalization helps improve the visualization of the image and makes details clearer.

[0177] Contrast Enhancement: If the contrast of a signal in an image is low, contrast enhancement methods (such as adaptive histogram equalization) can be used to further improve the signal's discernibility. Normalization: To enhance the contrast and brightness of an image, the reconstructed image can be normalized, standardizing the pixel values ​​to a suitable range to improve visualization.

[0178] This example presents a second harmonic signal reconstruction method based on amplitude-phase joint filtering. Through amplitude-phase joint filtering, this invention effectively removes noise from periodic signals while preserving their periodic characteristics. Traditional methods often only process the amplitude or phase of the signal, while this invention, by jointly processing amplitude and phase information, can more accurately reconstruct the target signal and improve imaging quality. Especially in low signal-to-noise ratio environments, the influence of noise is effectively suppressed, and the target signal is well preserved. It can be applied to the following fields:

[0179] First, this example can be applied to second harmonic microscopy systems, utilizing second harmonic signals generated by laser excitation for high-resolution imaging of tissue structures. Through denoising and signal enhancement, image quality can be significantly improved under low signal-to-noise ratio conditions, making it particularly suitable for observing tissue microstructures such as collagen fibers and extracellular matrix.

[0180] Secondly, this example can enhance the detection capability of biomolecules (such as DNA, RNA, and proteins) by improving signal quality when using fluorescent probes to label molecules, and can be applied to fields such as molecular biology and gene research.

[0181] In an optional embodiment, a second harmonic signal reconstruction process based on amplitude-phase joint filtering is used to reconstruct several original data images corresponding to each polarization angle, as follows: Figure 6 As shown. Furthermore, Figure 6 The structural strength indicated by the arrows and the strength at the box locations in the m original images are as follows: Figure 7A As shown; the structural strength indicated by the arrows and the strength at the box locations in the n enhanced images are as follows: Figure 7B As shown.

[0182] In an optional embodiment, the effect of this example is as follows: Figure 8 As shown.

[0183] This embodiment also provides a second harmonic signal reconstruction device based on amplitude and phase joint filtering. This device is used to implement the above embodiments and preferred embodiments, and will not be repeated for details already described. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0184] This embodiment provides a second harmonic signal reconstruction device based on amplitude and phase joint filtering, such as... Figure 9 As shown, the device includes:

[0185] The imaging module 801 is used to perform second harmonic polarization imaging on the sample using a microscopic imaging system to obtain multiple second harmonic intensity images, wherein the illumination polarization angle corresponding to each second harmonic intensity image is different.

[0186] Construction module 802 is used to construct a three-dimensional image stack based on the multiple second harmonic intensity images;

[0187] Transformation module 803 is used to perform Fourier transform on the three-dimensional image stack along the third dimension to obtain the target frequency domain dataset;

[0188] The filtering module 804 is used to perform joint filtering on the amplitude and phase of the target frequency domain data set, and reconstruct multiple initial time domain second harmonic intensity images after denoising.

[0189] The processing module 805 is used to perform adaptive threshold denoising processing on the multiple initial time-domain second harmonic intensity images to obtain multiple target time-domain second harmonic intensity images of the sample.

[0190] The second harmonic signal reconstruction apparatus based on amplitude-phase joint filtering provided in this embodiment of the invention can execute the second harmonic signal reconstruction method based on amplitude-phase joint filtering provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method. Further functional descriptions of the above modules are the same as in the corresponding embodiments described above, and will not be repeated here.

[0191] Figure 10 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0192] The following is a detailed reference. Figure 10The diagram illustrates a structural schematic suitable for implementing an electronic device according to embodiments of the present invention. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 901, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 902 or a program loaded from memory 908 into random access memory (RAM) 903. The RAM 903 also stores various programs and data required for the operation of the electronic device. The processor 901, ROM 902, and RAM 903 are interconnected via a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0193] Typically, the following devices can be connected to I / O interface 905: input devices 906 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 907 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 908 including, for example, magnetic tapes, hard disks, etc.; and communication devices 909. Communication device 909 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 10 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0194] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 909, or installed from a memory 908, or installed from a ROM 902. When the computer program is executed by the processor 901, it performs the functions defined in the second harmonic signal reconstruction method based on amplitude-phase joint filtering according to embodiments of the present invention.

[0195] Figure 10 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0196] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the second harmonic signal reconstruction method based on amplitude-phase joint filtering shown in the above embodiments is implemented.

[0197] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0198] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for reconstructing a second harmonic signal based on joint amplitude and phase filtering, characterized in that, The method includes: The sample was subjected to second harmonic polarization imaging using a microscopic imaging system, resulting in multiple second harmonic intensity images. The illumination polarization angle corresponding to each second harmonic intensity image was different. A three-dimensional image stack is constructed based on the multiple second harmonic intensity images; The three-dimensional image stack is subjected to a Fourier transform along the third dimension to obtain the target frequency domain dataset; The amplitude and phase of the target frequency domain dataset are jointly filtered, and multiple initial time-domain second harmonic intensity images are reconstructed after denoising. Adaptive threshold denoising is performed on the multiple initial time-domain second harmonic intensity images to obtain multiple target time-domain second harmonic intensity images of the sample. The target frequency domain dataset undergoes joint filtering of its amplitude and phase, and is reconstructed to obtain multiple initial time-domain second harmonic intensity images after denoising, including: The amplitude and phase of the target frequency domain data are jointly filtered to obtain multiple frequency domain signals; Perform inverse Fourier transform and time-domain reconstruction on the multiple frequency domain signals to obtain multiple time-domain signals; Based on the multiple time-domain signals, determine the multiple initial time-domain second harmonic intensity images after denoising; The process involves jointly filtering the amplitude and phase of the target frequency domain data to obtain multiple frequency domain signals. This includes: extracting the amplitude and phase of each spatial pixel from the target frequency domain data to obtain an amplitude matrix and a phase matrix; applying Gaussian filtering to the amplitude matrix and phase matrix respectively, and applying Gaussian filtering to the real and imaginary parts of the frequency domain signals respectively; and recombinating the filtered amplitude and phase into multiple frequency domain signals in complex form.

2. The method according to claim 1, characterized in that, Perform a Fourier transform on the three-dimensional image stack along the third dimension to obtain the target frequency domain dataset, including: The three-dimensional image stack is subjected to a Fourier transform along the third dimension to obtain the initial frequency domain dataset; The fundamental frequency component is extracted from the initial frequency domain dataset to obtain the target frequency domain dataset.

3. The method according to claim 2, characterized in that, The initial frequency domain dataset is subjected to fundamental frequency component extraction to obtain the target frequency domain dataset, which includes: Obtain the polarization angle modulation period value; Based on the polarization angle modulation period value and frequency domain conjugate symmetry, the fundamental frequency component is extracted from the initial frequency domain dataset to obtain the target frequency domain dataset.

4. The method according to claim 1, characterized in that, Adaptive threshold denoising is performed on the multiple initial time-domain second harmonic intensity images to obtain multiple target time-domain second harmonic intensity images of the sample, including: Gray-level histogram analysis was performed on the multiple initial time-domain second harmonic intensity images, and an adaptive threshold was determined; Based on the adaptive threshold, the multiple initial time-domain second harmonic intensity images are denoised to obtain the multiple target time-domain second harmonic intensity images of the sample.

5. The method according to claim 1, characterized in that, The method further includes: The time-domain second harmonic intensity images of the multiple targets are normalized and contrast-enhanced.

6. The method according to claim 1, characterized in that, The method further includes: The polarization orientation of the sample is determined based on the multiple target time-domain second harmonic intensity images.

7. A second harmonic signal reconstruction device based on amplitude and phase joint filtering, characterized in that, The device includes: The imaging module is used to perform second harmonic polarization imaging on the sample using a microscopic imaging system to obtain multiple second harmonic intensity images, wherein the illumination polarization angle corresponding to each second harmonic intensity image is different. The construction module is used to construct a three-dimensional image stack based on the multiple second harmonic intensity images; The transformation module is used to perform a Fourier transform on the three-dimensional image stack along the third dimension to obtain the target frequency domain dataset; The filtering module is used to perform joint filtering on the amplitude and phase of the target frequency domain dataset and reconstruct multiple initial time-domain second harmonic intensity images after denoising. The processing module is used to perform adaptive threshold denoising processing on the multiple initial time-domain second harmonic intensity images to obtain multiple target time-domain second harmonic intensity images of the sample. Specifically, the filtering module is used to: perform joint filtering on the amplitude and phase of the target frequency domain data to obtain multiple frequency domain signals; perform inverse Fourier transform and time-domain reconstruction on the multiple frequency domain signals to obtain multiple time-domain signals; and determine the denoised multiple initial time-domain second harmonic intensity images based on the multiple time-domain signals. The process involves jointly filtering the amplitude and phase of the target frequency domain data to obtain multiple frequency domain signals. This includes: extracting the amplitude and phase of each spatial pixel from the target frequency domain data to obtain an amplitude matrix and a phase matrix; applying Gaussian filtering to the amplitude matrix and phase matrix respectively, and applying Gaussian filtering to the real and imaginary parts of the frequency domain signals respectively; and recombinating the filtered amplitude and phase into multiple frequency domain signals in complex form.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to execute the second harmonic signal reconstruction method based on amplitude and phase joint filtering as described in any one of claims 1 to 6.

9. A computer program product, characterized in that, Includes computer instructions for causing a computer to execute the second harmonic signal reconstruction method based on amplitude-phase joint filtering as described in any one of claims 1 to 6.

Images