A confocal microscopy imaging method based on bidirectional scanning

Abstract

The present invention discloses a confocal microscopy imaging method based on bidirectional scanning. First, the control signal of the fast-axis galvanometer is replaced by a variable-amplitude cosine wave from a traditional triangle wave to avoid the jitter problem caused by sudden stops and turns. Then, in view of the misalignment of image pixel positions, the galvanometer step time is adaptively corrected using a symmetry matching algorithm to obtain imaging step pixels for real-time imaging. Then, in view of the nonlinear distortion of the scanning galvanometer and the deformation caused by the cosine wave scanning method collected by the acquisition system at equal time intervals, the pixel coordinates are remapped through velocity field modeling to eliminate the deformation. Using the variable-amplitude cosine wave as the control signal of the fast-axis galvanometer can make the movement of the scanning galvanometer smooth and continuous, avoid the occurrence of high-frequency howling, and increase the service life of the scanning galvanometer. The proportion of effective data in the bidirectional scanning of the present invention is increased to more than 95%, and the frame rate is increased by 2.4 times compared with the traditional unidirectional scanning at a resolution of 512×512, and the imaging efficiency is doubled.

Description

Technical Field

[0001] The invention relates to a confocal microscopic imaging method, in particular to a confocal microscopic imaging method based on bidirectional scanning. Background Art

[0002] Confocal scanning imaging technology, a crucial component of modern microscopy systems, relies on high-precision deflection and modulation of the illumination spot on the sample plane through a two-dimensional galvanometer module, enabling point-by-point scanning within the field of view. During this process, photomultiplier tubes detect the fluorescence signal emitted by the sample to generate pixel illumination values. This is then combined with a synchronously controlled data acquisition system to generate a two-dimensional light intensity distribution map, ultimately constructing a high-resolution image. However, traditional scanning control strategies still face significant technical bottlenecks in improving imaging efficiency and ensuring image quality.

[0003] In a classic raster scanning architecture, 2D galvanometer modules typically employ a unidirectional scanning mode: valid data acquisition occurs during the positive X-axis travel phase, while the reverse travel phase, used only for galvanometer reset and ineffective data acquisition, occurs. This system effectively utilizes less than 45% of the scan cycle, resulting in over 55% of time wasted in the mechanical reset process. With the growing demand for high-speed dynamic imaging (e.g., in real-time observation of living cells and industrial online inspection), this mode has become a key limiting factor in improving the temporal resolution of microscopy systems.

[0004] The bidirectional scanning scheme proposed in recent years uses a forward and reverse bidirectional acquisition strategy, which theoretically can improve scanning efficiency to 1.8-2 times that of traditional methods. However, in practical applications, the following technical difficulties have not been fundamentally resolved: (1) nonlinear distortion of the scanning path caused by the asymmetry of the mechanical response during the bidirectional movement of the galvanometer; (2) pixel position misalignment caused by rapid reversal; (3) inconsistency between the reverse stroke signal and the forward stroke dynamic characteristics, forcing the system to only capture the middle 30%-50% of the scanning trajectory as valid data. This not only greatly reduces the scanning efficiency (actually improved by about 40%-60%), but also causes information loss in the edge areas of the image. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a confocal microscopic imaging method based on bidirectional scanning with high scanning efficiency and more complete information.

[0006] The technical solution adopted by the present invention to solve the above technical problems is: a confocal microscopy imaging method based on bidirectional scanning, comprising the following steps:

[0007] Step 1: The control system uses the variable amplitude cosine wave as a control signal for the fast-axis galvanometer of the scanning galvanometer, and uses the step wave as a control signal for the slow-axis galvanometer in the control signal of the scanning galvanometer, wherein the rising period of the step wave is equal to half the period of the variable amplitude cosine wave, and each voltage step of the step wave corresponds to the scanning galvanometer scanning a line of the image;

[0008] Step 2: When the control system inputs a control signal to start the movement of the scanning galvanometer, the acquisition system acquires the light intensity signal fed back by the photomultiplier tube at equal time intervals;

[0009] Step 3: Preliminarily reconstructing a two-dimensional image matrix based on the original signal stream of the light intensity signal acquired by the acquisition system, wherein the first half of each row of the two-dimensional image matrix is ​​the data scanned by the fast-axis galvanometer of the scanning galvanometer, and the second half of each row of the two-dimensional image matrix is ​​the data scanned by the fast-axis galvanometer of the scanning galvanometer;

[0010] Step 4: Cut data of different lengths from the two-dimensional image matrix to construct several sub-matrices, compare the symmetry matching of the data scanned by the fast-axis galvanometer in the forward stroke and the data scanned by the fast-axis galvanometer in the reverse stroke, and find the imaging step pixels;

[0011] Step 5: Taking the first-order derivative of the control signal of the fast-axis galvanometer of the scanning galvanometer described in step 1 to obtain an instantaneous velocity sinusoidal curve;

[0012] Step 6: Discretize the instantaneous velocity sinusoidal curve into N grids, where N is the resolution of the fast-axis galvanometer scanning direction of the scanning galvanometer, extract the velocities of the N nodes, and generate a velocity set;

[0013] Step 7: Construct a two-dimensional image matrix of the original signal stream of the light intensity signal based on the imaging step pixels obtained in step 4, and remap the pixel coordinates according to the velocity set to eliminate the cosine waveform variation and obtain the final two-dimensional confocal microscopy image.

[0014] Compared with the existing technology, the advantage of the present invention is that it breaks through the long-standing "speed-precision" trade-off dilemma in the field of confocal microscopy imaging. First, the control signal of the fast-axis galvanometer of the scanning galvanometer is replaced by a variable-amplitude cosine wave from the traditional triangle wave to avoid the jitter problem caused by sudden stops and turns. Then, in response to the image pixel position misalignment phenomenon, the galvanometer step time is adaptively corrected using a symmetry matching algorithm to obtain imaging step pixels for real-time imaging. Then, in response to the nonlinear distortion of the scanning galvanometer and the deformation caused by the cosine wave scanning method collected by the acquisition system at equal time intervals, the pixel coordinates are remapped through velocity field modeling to eliminate the deformation. The use of a variable-amplitude cosine wave as the control signal of the fast-axis galvanometer can make the movement of the scanning galvanometer smooth and continuous, avoid the occurrence of high-frequency howling, and increase the service life of the scanning galvanometer. The effective data ratio of the bidirectional scanning of the present invention is increased to more than 95%, and the frame rate is increased by 2.4 times compared with the traditional unidirectional scanning at a resolution of 512×512, and the imaging efficiency is doubled.

[0015] Specifically, the variable amplitude cosine wave in step 1 is expressed as V = A cos(2π ft ), where A is the dynamic amplitude modulation coefficient, which is used to control the scanning range of the scanning galvanometer, π is the pi, f is the frequency of the variable amplitude cosine wave, which is used to control the fast axis motion frequency of the scanning galvanometer. t is the time variable.

[0016] Preferably, the imaging step pixel in step 4 refers to the lag data length between the pixel position and the actual position caused by the step response time of the scanning galvanometer.

[0017] Preferably, the two-dimensional image matrix construction described in step 7 refers to shifting the data within the effective data interception range of the original signal stream of the light intensity signal backward by the imaging step pixel data length to obtain a non-misaligned confocal image.

[0018] Preferably, eliminating the cosine waveform variation in step 7 refers to correcting the misaligned confocal image according to the speed set, taking the speed at 1 / 4 period of the variable amplitude cosine wave as the standard speed, merging the excess sampling points in the low-speed area, and expanding the missing sampling points in the high-speed area to obtain an image without cosine waveform variation. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a flowchart of a confocal microscopy imaging method based on bidirectional scanning according to an embodiment of the present invention;

[0020] Figure 2is the waveform of the variable amplitude cosine wave of the control signal of the fast axis galvanometer in the embodiment of the present invention;

[0021] Figure 3 is the waveform of the step wave of the control signal of the slow-axis galvanometer in the embodiment of the present invention;

[0022] Figure 4 Schematic diagram of the scanning path of the scanning galvanometer in an embodiment of the present invention;

[0023] Figure 5 3. This is a schematic diagram of a process of shifting data within a valid data interception range of an original signal stream of a light intensity signal forward by A data lengths in an embodiment of the present invention;

[0024] Figure 6 A two-dimensional image with a resolution of 1024×256 is constructed using the original signal stream of the light intensity signal in step 4 of the embodiment of the present invention;

[0025] Figure 7 It is a two-dimensional image matrix preliminarily reconstructed based on the original signal stream of the light intensity signal acquired by the acquisition system in the embodiment of the present invention;

[0026] Figure 8 is a two-dimensional image having a cosine waveform in an embodiment of the present invention;

[0027] Figure 9 This is the two-dimensional confocal microscopic image finally obtained in the embodiment of the present invention. DETAILED DESCRIPTION

[0028] The present invention will be described in further detail below with reference to the accompanying drawings and embodiments.

[0029] Example 1: A confocal microscopy imaging method based on bidirectional scanning, such as Figure 1 As shown, the following steps are included:

[0030] Step 1: Scanning galvanometer control signal configuration:

[0031] The control system replaces the control signal of the fast axis galvanometer of the traditional scanning galvanometer with Figure 2 The variable amplitude cosine wave with a period of T is shown; the variable amplitude cosine wave is expressed as V = A cos(2π ft ), where A is the dynamic amplitude modulation coefficient, which is used to control the scanning range of the scanning galvanometer, π is the pi, f is the frequency of the variable amplitude cosine wave, which is used to control the fast axis motion frequency of the scanning galvanometer. t is the time variable;

[0032] The control system sets the control signal of the slow axis galvanometer in the control signal of the scanning galvanometer as follows Figure 3 The rising period of the step wave shown is T / 2, and the center of its rising edge is strictly aligned with the symmetry axis of the variable amplitude cosine wave.

[0033] Step 2: Scan timing planning:

[0034] Taking scanning a 0.1mm grid plate as an example, the control system inputs a control signal to make the scanning galvanometer scan an image with a resolution of 512×512. The scanning path is as follows: Figure 4 As shown, where:

[0035] The fast-axis galvanometer of the scanning galvanometer performs periodic reciprocating scanning according to the variable-amplitude cosine wave. One cosine wave cycle represents one reciprocating motion completed by the fast-axis galvanometer, that is, the fast-axis galvanometer moves 512×2 points in the horizontal direction.

[0036] The slow-axis galvanometer of the scanning galvanometer realizes step-by-step displacement according to the step wave. One step represents that the slow-axis galvanometer of the scanning galvanometer moves one point in the vertical direction, while the fast-axis galvanometer moves 512 points in the horizontal direction.

[0037] Step 3: Synchronize data acquisition:

[0038] When the control system inputs a control signal to start the scanning galvanometer movement, the acquisition system is triggered to collect the light intensity signal fed back by the photomultiplier tube at equal time intervals to ensure that the timestamps are aligned.

[0039] Step 4: Original image construction:

[0040] The control system controls the scanning galvanometer to complete Figure 4 When scanning the path in the image, the acquisition system simultaneously collects the position information of 512×512 light intensity signals, and constructs the original signal flow of the light intensity signal composed of the position information of these light intensity signals as follows: Figure 6 The two-dimensional image shown has a resolution of 1024×256 and features include:

[0041] (1) The first 512 points in each row are the data of the fast-axis galvanometer of the scanning galvanometer during the positive stroke (right direction);

[0042] (2) The last 512 points in each row are the data of the reverse scan of the fast axis galvanometer of the scanning galvanometer (left direction);

[0043] Since the scanning mirror moves and collects data simultaneously in step 3, the scanning mirror has a step response time (movement time), and the collected position information is not the movement destination position of the scanning mirror (typically manifested as non-central symmetric distortion, see Figure 7 ).

[0044] Step 5: Dynamic Delay Calibration:

[0045] The original signal stream of the light intensity signal is shifted forward by 1, 2, 3, ..., 512 data lengths respectively to generate Figure 5 The new 512 data streams shown are filled with black at the end to construct 512 two-dimensional images with a resolution of 1024×256;

[0046] The forward / reverse similarity of each two-dimensional image is calculated based on the horizontal symmetry axis of the two-dimensional image, and the lag data length between the pixel position corresponding to the maximum similarity value and the actual position is selected as the imaging step pixel A.

[0047] Step 6: Non-uniform scanning distortion correction:

[0048] After the data in the effective data interception range of the original signal stream of the light intensity signal is shifted forward by A data length, the reconstruction is as follows: Figure 8 The resolution of the two-dimensional image shown is 512×512. Due to the non-uniform horizontal motion of the scanning mirror and the acquisition system collecting data at equal time intervals, the reconstructed two-dimensional image has cosine wave deformation in the horizontal direction. Therefore, the horizontal cosine waveform correction is performed based on the instantaneous velocity sine curve of the fast axis galvanometer of the scanning mirror:

[0049] (1) Take the first-order derivative of the control signal of the fast-axis galvanometer of the scanning galvanometer to obtain the instantaneous velocity sine curve;

[0050] (2) Discretize the instantaneous velocity sinusoid into 512 grid nodes and generate the velocity set V = {v1,...,v n ,...,v N}, where n represents the speed number of the node, and the value range of n is 1~512, N=512;

[0051] (3) Taking the speed at 1 / 4 of the cosine wave of the control signal of the fast-axis galvanometer as the standard speed, merge the excessive sampling points in the low-speed area (oversampling), expand the missing sampling points in the high-speed area (undersampling), and obtain the following Figure 9 The image shown is a two-dimensional confocal microscopy image without cosine waveform deformation.

[0052] Confocal scanning imaging uses a cosine wave voltage control signal to control the scanning of the scanning galvanometer. When scanning an image with a resolution of 512×512, the frame rate can reach 1.89 frames / s. Compared with the 0.79 frames / s scanned using a triangle wave voltage control signal, the scanning speed is increased by 2.4 times, and the effective utilization rate of the scanning cycle reaches 98%.

Claims

1. A confocal microscopy imaging method based on bidirectional scanning, characterized in that The following steps are involved: Step 1: The control system uses the variable amplitude cosine wave as the control signal of the fast-axis galvanometer in the control signal of the scanning galvanometer, and uses the step wave as the control signal of the slow-axis galvanometer in the control signal of the scanning galvanometer, wherein the rising period of the step wave is equal to half the period of the variable amplitude cosine wave, and each voltage step of the step wave corresponds to the scanning galvanometer scanning a line of the image; Step 2: When the control system inputs a control signal to start the movement of the scanning galvanometer, the acquisition system acquires the light intensity signal fed back by the photomultiplier tube at equal time intervals; Step 3: Preliminarily reconstructing a two-dimensional image matrix based on the original signal stream of the light intensity signal acquired by the acquisition system, wherein the first half of each row of the two-dimensional image matrix is ​​the data scanned by the fast-axis galvanometer of the scanning galvanometer, and the second half of each row of the two-dimensional image matrix is ​​the data scanned by the fast-axis galvanometer of the scanning galvanometer; Step 4: Cut data of different lengths from the two-dimensional image matrix to construct several sub-matrices, compare the symmetry matching of the data scanned by the fast-axis galvanometer in the forward stroke and the data scanned by the fast-axis galvanometer in the reverse stroke, and find the imaging step pixels; Step 5: Taking a first-order derivative of the control signal of the fast-axis galvanometer in the control signal of the scanning galvanometer described in step 1 to obtain an instantaneous velocity sine curve; Step 6: Discretize the instantaneous velocity sinusoidal curve into N grids, where N is the resolution of the fast-axis galvanometer scanning direction of the scanning galvanometer, extract the velocities of the N nodes, and generate a velocity set; Step 7: Construct a two-dimensional image matrix of the original signal stream of the light intensity signal based on the imaging step pixels obtained in step 4, and remap the pixel coordinates according to the velocity set to eliminate the cosine waveform variation and obtain the final two-dimensional confocal microscopy image.

2. A confocal microscopy imaging method based on bidirectional scanning according to claim 1, characterized in that The variable amplitude cosine wave in step 1 is expressed as V = A cos(2π ft ), where A is the dynamic amplitude modulation coefficient, which is used to control the scanning range of the fast axis galvanometer, π is the pi, f is the frequency of the variable amplitude cosine wave, which is used to control the movement frequency of the fast axis galvanometer. t is the time variable.

3. A confocal microscopy imaging method based on bidirectional scanning according to claim 1, characterized in that The imaging step pixel mentioned in step 4 refers to the lag data length between the pixel position and the actual position caused by the step response time of the scanning galvanometer.

4. A confocal microscopy imaging method based on bidirectional scanning according to claim 3, characterized in that The two-dimensional image matrix construction described in step 7 refers to shifting the data within the effective data interception range of the original signal stream of the light intensity signal backward by the imaging step pixel data length to obtain a non-misaligned confocal image.

5. A confocal microscopy imaging method based on bidirectional scanning according to claim 4, characterized in that Eliminating the cosine waveform variation in step 7 refers to correcting the misaligned confocal image according to the speed set, taking the speed at 1 / 4 period of the variable amplitude cosine wave as the standard speed, merging the excess sampling points in the low-speed area, and expanding the missing sampling points in the high-speed area to obtain an image without cosine waveform variation.

Images