OCT retinal partition brightness self-adaptive clear imaging adjustment method and system

Abstract

The present application belongs to the field of OCT imaging technology, and particularly relates to an OCT retinal partition brightness adaptive clear imaging adjustment method and system. The method comprises the following steps: determining five regions including the center and four corners of an OCT retinal image; extracting the brightness features of the five regions; determining the relative position of the OCT retinal image spot and the pupil based on the relationship between the brightness features of the five regions and the preset threshold; and constructing a corresponding error function according to the relative position of the OCT retinal image spot and the pupil, and adjusting the spot position by using a PID control method. The brightness feature analysis dimension of the present application is the first to combine the center and four-quadrant partition statistics with a weighted mean algorithm, and the pupil alignment degree is classified and judged according to the spot state, thereby providing accurate basis for dynamic adjustment. The error function of proportional compensation and differential control is introduced to balance the response speed and stability of the adjustment, and the adjustment efficiency and the stability of the imaging quality are improved through pupil position prediction and reverse adjustment logic.

Description

Technical Field

[0001] This invention belongs to the field of OCT imaging technology, and particularly relates to an OCT retinal partition brightness adaptive clear imaging adjustment method and system. Background Technology

[0002] Optical coherence tomography (OCT) is a technique that uses the principle of light interference to achieve non-contact three-dimensional detection. By capturing the reflected and backscattered light signals of a sample tissue, this technique can accurately reconstruct cross-sectional and three-dimensional structural images of the tissue. With its advantages of high resolution, non-invasiveness, high real-time performance, and high sensitivity, it has become the gold standard for the diagnosis of ophthalmic diseases.

[0003] Currently, in the process of OCT equipment testing the human eye, determining the position of the pupil and whether the incident light spot correctly enters the pupil range are key steps to achieve automatic adjustment and image clarity. However, existing technologies still have several limitations in this process.

[0004] Existing OCT automatic adjustment technologies mainly rely on simple threshold segmentation of pupil camera images, continuous frame cross-correlation, or a single sharpness index to determine the position of the light spot and the imaging state. However, these methods are easily affected by factors such as corneal reflection, tear film reflection, blurred pupil edges, and uneven field of view brightness. They cannot accurately identify whether the light spot has completely entered the pupil, nor can they determine the direction and amount of offset. At the same time, the automatic control strategies of existing OCT systems mostly rely on empirical thresholds, intelligently selecting between overall device movement and reference arm adjustment. This results in low automation and poor adjustment efficiency, making it difficult to image in patients with low cooperation during testing. Summary of the Invention

[0005] To overcome the shortcomings of the prior art, this invention provides an OCT retinal zone brightness adaptive clear imaging adjustment method and system. By analyzing the brightness distribution of each azimuth angle in the OCT field of view and combining it with the clarity index of the pupil camera, the brightness of each azimuth angle and the center during OCT retinal imaging is detected. This enables automatic judgment of whether the light spot has completely entered the pupil range, and the overall movement of the device and the reference arm is adjusted according to the judgment result. This allows for automatic closed-loop imaging optimization from coarse adjustment to fine adjustment, achieving clear imaging.

[0006] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions:

[0007] The first aspect of this invention provides an OCT retinal partition brightness adaptive clear imaging adjustment method.

[0008] The OCT retinal zone brightness adaptive clear imaging adjustment method includes the following steps:

[0009] Acquire an OCT retinal image and determine five regions in total, including the center and four corners of the OCT retinal image;

[0010] The brightness features of five regions are extracted, including the average brightness of the central region, the average brightness of the four corner regions, the global average brightness, the average brightness of the four corners, and the ratio of the four corners to the center.

[0011] Based on the relationship between the brightness characteristics of five regions and preset thresholds, the relative positions of the light spots and pupils in the OCT retinal image are determined.

[0012] Based on the relative position of the light spot and the pupil in the OCT retinal image, a corresponding error function is constructed. The PID control method is used to track the target brightness values ​​of five regions, thereby adjusting the position of the light spot and enabling the OCT retinal imaging to achieve the preset clarity.

[0013] A second aspect of the present invention provides an OCT retinal partition brightness adaptive clear imaging adjustment system.

[0014] The OCT retinal zone brightness adaptive clear imaging adjustment system includes:

[0015] The multi-region determination module is configured to: acquire OCT retinal images and determine five regions, including the center and four corners of the OCT retinal images;

[0016] The brightness feature extraction module is configured to extract brightness features of five regions, including the average brightness of the central region, the average brightness of the four corner regions, the global average brightness, the average brightness of the four corners, and the ratio of the four corners to the center.

[0017] The spot location module is configured to determine the relative position of the spot in the OCT retinal image and the pupil based on the relationship between the brightness characteristics of five regions and a preset threshold.

[0018] The PID tracking and adjustment module is configured to: construct a corresponding error function based on the relative position of the light spot and the pupil in the OCT retinal image, and use the PID control method to track the target brightness values ​​of five regions to adjust the position of the light spot so that the OCT retinal imaging achieves the preset clarity.

[0019] A third aspect of the present invention provides a computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps of the OCT retinal partition brightness adaptive clear imaging adjustment method as described in the first aspect of the present invention.

[0020] The fourth aspect of the present invention provides an electronic device, including a memory, a processor, and a program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the OCT retinal partition brightness adaptive clear imaging adjustment method as described in the first aspect of the present invention.

[0021] The above one or more technical solutions have the following beneficial effects:

[0022] This invention provides an OCT retinal partition brightness adaptive clear imaging adjustment method and system. The brightness feature analysis dimension innovatively combines center + four-quadrant partition statistics with a weighted average algorithm to classify and determine pupil alignment based on the light spot state, providing a precise basis for dynamic adjustment. In the control strategy part, error functions of proportional compensation and differential control are introduced to balance the adjustment response speed and stability. At the same time, through pupil position prediction and reverse adjustment logic, the adjustment efficiency and image quality stability are improved, ultimately realizing the device's adaptive high-definition imaging for different pupils.

[0023] This invention calculates five brightness feature values: average brightness of the central region, average brightness of the four corner regions, global average brightness, average brightness of the four corners, and the ratio of the four corners to the center. It also combines these with preset criteria to determine the relative position of the current light spot and the pupil, resulting in three determination results. The pupil alignment is determined based on the light spot state, providing a precise basis for PID dynamic adjustment.

[0024] When determining whether a light spot is off-center or tilted, this invention incorporates X-direction and Y-direction errors into the error function. By normalizing and weighting the brightness difference between a single-sided region and the combined region, it can effectively distinguish between the overall light spot offset and the interference of local brightness noise, while also eliminating the influence of the absolute value of the overall brightness, directly outputting the relative offset degree in the X and Y directions. Furthermore, it can directly correspond to the offset direction and accurately input it into the error function for subsequent calculations.

[0025] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0026] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0027] Figure 1 This is a flowchart of the method in Example 1.

[0028] Figure 2 This is a schematic diagram of the OCT retinal imaging adaptive adjustment system in Example 1.

[0029] Figure 3 This is the overall optical path diagram of the reference arm adjustment unit in Embodiment 1.

[0030] Figure 4 This is a schematic diagram showing the light spot at the fixed point of Example 1 completely entering the pupil.

[0031] Figure 5 This is a schematic diagram showing that the light spot at the fixed point in Example 1 did not completely enter the pupil.

[0032] Figure 6 This is a schematic diagram of the locations of the five regions in Example 1.

[0033] Figure 7 The flowchart for brightness status judgment and PID adjustment in Example 1 is shown.

[0034] The attached diagram lists the components represented by each number as follows:

[0035] 1. First lens; 2. Second lens; 3. Optical axis; 4. Pupil; 5. Iris; 6. Fixed point; 7. Collimator; 8. First reflecting mirror; 9. Second reflecting mirror; 10. Third reflecting mirror; 11. Fourth reflecting mirror; 12. Dichroic mirror; 13. Visible light; 14. Infrared light; 15. Light spot at the fixed point; 16. Scanning area; 17. Central area; 18. Upper left area; 19. Upper right area; 20. Lower left area; 21. Lower right area; 22. Focal point. Detailed Implementation

[0036] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0037] It should be noted that the terminology used herein is for the purpose of describing particular implementations only and is not intended to limit the exemplary implementations of the present invention.

[0038] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0039] Example 1

[0040] As mentioned earlier, existing OCT systems suffer from various problems such as inaccurate pupil positioning during automatic adjustment, leading to difficulties in image optimization and unstable image quality. Therefore, this embodiment discloses an OCT retinal zone brightness adaptive clear imaging adjustment method. First, five regions are determined: the center of the OCT field of view and each azimuth angle. The brightness distribution of these five regions is obtained, and brightness characteristics are calculated. Based on the relationship between the brightness characteristics and a preset threshold, the relative position of the OCT retinal image spot and the pupil is determined. This enables automatic judgment of whether the spot has completely entered the pupil range. Based on the judgment result, an error function is constructed, and combined with PID control, the overall movement of the device and the reference arm is adjusted. This achieves automatic closed-loop imaging optimization of the OCT system from coarse to fine adjustment, resulting in high-quality OCT retinal imaging.

[0041] like Figure 1 As shown, the OCT retinal partition brightness adaptive clear imaging adjustment method includes the following steps:

[0042] Acquire an OCT retinal image and determine five regions in total, including the center and four corners of the OCT retinal image;

[0043] The brightness features of five regions were extracted, including the average brightness of the central region, the average brightness of the four corner regions, the global average brightness, the average brightness of the four corner regions, and the ratio of the four corner regions to the center.

[0044] Based on the relationship between the brightness characteristics of five regions and preset thresholds, the relative positions of the light spots and pupils in the OCT retinal image are determined.

[0045] Based on the relative position of the light spot and the pupil in the OCT retinal image, a corresponding error function is constructed. The PID control method is used to track the target brightness values ​​of five regions, thereby adjusting the position of the light spot and enabling the OCT retinal imaging to achieve the preset clarity.

[0046] This embodiment considers the brightness of five regions, including the center of the OCT field of view and the various azimuth angles of the field of view. It can be understood that the four corner regions of the OCT retinal image are the upper left, upper right, lower left, and lower right corners.

[0047] When extracting the brightness features of the five regions, a total of five feature values ​​were calculated. The calculation method is summarized as follows:

[0048] Calculate the average pixel brightness of the central region 17 to obtain the average brightness of the central region 17;

[0049] Calculate the average pixel brightness of the four corner regions to obtain the average brightness of the four corner regions;

[0050] The global average brightness is calculated by combining the weighted average of the average brightness of the central region (17) and the average brightness of the four corner regions.

[0051] Set a normalization term for the brightness difference between the center and the corners, and calculate the average brightness of the four corners by weighting the brightness difference normalization term between the center and the corners.

[0052] The ratio of the average brightness of the four corner regions to the average brightness of the central region 17 is used to calculate the ratio of the brightness difference between the four corner regions relative to the center by averaging the normalized difference terms of the brightness difference between the central region 17 and the average brightness of the four corner regions.

[0053] It can be understood that the average brightness of the four corner areas and the average brightness of the four corners calculated above are two different concepts.

[0054] After obtaining the brightness characteristics of the above five regions, the relative position of the light spot and the pupil in the OCT retinal image can be determined based on the preset criteria (i.e., preset threshold). Different relative positions of the light spot and the pupil will constitute different states. After determining the states of the two, a corresponding error function is constructed in combination with the specific states, and the PID control method is used to adjust the movement of the actuator (the whole device and the reference arm) to achieve the purpose of clear and high-quality imaging.

[0055] Overall, the innovation of the OCT retinal partition brightness adaptive clear imaging adjustment method provided in this embodiment focuses on two aspects:

[0056] First, the analysis dimension is the first to combine the center + four-quadrant partition statistics with a weighted mean algorithm, which classifies and judges the pupil alignment based on the light spot state, providing a precise basis for dynamic adjustment;

[0057] Secondly, the control strategy introduces error functions of proportional compensation and differential control to balance the adjustment response speed and stability. At the same time, through pupil position prediction and reverse adjustment logic, the adjustment efficiency and image quality stability are improved, ultimately realizing the device's adaptive high-definition imaging for different pupils.

[0058] The technical solution of this embodiment will now be explained in detail with reference to the accompanying drawings.

[0059] (I) System Setup.

[0060] The OCT retinal partition brightness adaptive clear imaging adjustment method provided in this embodiment relies on an existing OCT retinal imaging adaptive adjustment system. The OCT retinal imaging adaptive adjustment system uses the OCT host as the core imaging unit, providing a low-coherence light source, interferometer, and signal acquisition functions to generate retinal interference signals and convert them into image data.

[0061] The equipment composition and system architecture logic of the OCT retinal imaging adaptive adjustment system are as follows:

[0062] Figure 2This is a schematic diagram of the OCT retinal imaging adaptive adjustment system in this embodiment. Figure 2 As shown, two lenses with matching focal lengths (first lens 1 and second lens 2) are coaxially mounted to form a 4f optical system;

[0063] The positions of the iris 5 and pupil 4 of a person are shown in the direction of the second lens 2 away from the first lens 1;

[0064] The incident parallel light is focused along the optical axis 3 by the first lens 1 to the first focal point 22, and then collimated into parallel light by the second lens 2, and finally focused on the fixed point 6 in the region of the human eye pupil 4.

[0065] The entire 4f optical system and scanning galvanometer are integrated into the sample arm assembly. The sample arm assembly is translated along the optical axis 3 by a servo motor slide to adjust the relative position of the fixed point 6 and the pupil 4.

[0066] The side-mounted pupil camera acquires pupil images in real time through a conjugate optical path to provide positional deviation feedback, while the reference arm adjustment unit performs optical path compensation through a micro-displacement platform, working in conjunction with the sample arm assembly to complete closed-loop imaging optimization.

[0067] like Figure 3 This is the overall optical path diagram of the reference arm adjustment unit. (Example:) Figure 3 As shown, the first reflector 8 and the second reflector 9 are placed on the motor slide, and the optical path of the reference arm can be adjusted by controlling the motor.

[0068] The light beam output from the collimator 7 is reflected in sequence by the first reflector 8 and the second reflector 9, and then incident on the dichroic mirror 12. The first reflector 8 and the second reflector 9 are moved back and forth by a motor to adjust the optical path of the reference arm.

[0069] The dichroic mirror 12 separates the visible light 13 and the infrared light 14. The visible light 13 is reflected to the third mirror 10, while the infrared light 14 is transmitted to the fourth mirror 11. Then the two beams return along the original optical path and interfere with the light from the sample arm.

[0070] Each reflector and dichroic mirror 12 is precisely installed at a preset angle to ensure the separation and accurate guidance of the light paths of visible light 13 and infrared light 14.

[0071] The OCT retinal imaging adaptive adjustment system used in this embodiment is an existing mature technology, and the specific setting principle and adjustment details will not be elaborated further.

[0072] (II) System initialization.

[0073] In actual testing, the pupil size, eye position and refractive state of different individuals vary greatly, which causes the focal point 22 (fixed point 6) of the incident light to not completely coincide with the position of the pupil, resulting in problems such as blurred imaging, uneven field of view and decreased signal-to-noise ratio.

[0074] In the initial stage of OCT equipment actually examining a patient's eyeball, because everyone's eyeball and pupil size are different, the pupil camera may not be able to find a clear pupil image, which means that light is not entering the pupil, leading to imaging failure. Therefore, the first step is to align the pupil camera with the focal plane of the second lens 2 in the 4f optical system.

[0075] In the initial stage of data acquisition, the pupil camera is observed. When the pupil camera cannot locate the patient's pupil, the image will be very dark. At this point, the entire device needs to be moved to change the distance between the second lens 2 in the 4f optical system and the human pupil 4. The first step is to allow the pupil camera to locate the pupil 4. Considering the patient's safety, the device should be moved backward first (moving the device forward first might bump into the patient's face), while simultaneously moving the reference arm backward to increase the optical path. Once the outline of the human eye is detected, the device's vertical and horizontal orientation needs to be adjusted to align the pupil camera with the patient's pupil 4.

[0076] The next step is observation. If the pupil camera image becomes clearer when the device is moved backward, it indicates that the device is moving in the correct direction. In this case, simply move the device backward to the position where the pupil camera image is clearest. If a clear image cannot be obtained while moving the device backward, it means that the device is moving in the wrong direction. Therefore, a maximum backward distance needs to be set. If a clear image is still not found after moving the device to the maximum distance, then the device needs to be adjusted in the opposite direction.

[0077] In a 4f optical system, the pupil, as the target to be imaged, must be within the focal plane of the second lens 2 or its conjugate plane to achieve a clear image. When the adjustment distance exceeds half the focal length, the pupil will fall out of the lens's effective imaging range, and even reverse adjustment will not restore a clear image.

[0078] In the sample arm of the OCT device used in this embodiment, the focal length of the second lens 2 of the 4f optical system is f = 25 mm, so the maximum distance is set to l = f / 2 = 12.5 mm. This provides sufficient adjustment margin to accommodate the physiological differences of different patients, avoids mechanical errors or safety risks caused by excessive device movement, and shortens ineffective adjustment time, thereby improving detection efficiency.

[0079] Therefore, if a clear image is not obtained when the device is moved back by f / 2, the device is moved forward to the position where the pupil camera image is clearest.

[0080] After completing the coarse adjustment, the pupil camera is basically aligned with the center of the pupil. At this point, we enter the semi-fine adjustment stage, where we analyze the brightness characteristics of the OCT retinal image to further determine the relative position of the light spot and the pupil.

[0081] (III) Method for identifying the location of light spots.

[0082] After the second step is completed, the pupil camera is basically aligned with the pupil position. At this point, adjusting the reference arm to bring the OCT retinal image to the zero-frequency position allows for basic image formation, but the image quality is low, and the retinal image is unclear. This is because the light spot at fixed point 6 has not yet truly overlapped with pupil 4, requiring semi-fine adjustment.

[0083] During semi-fine adjustment, the brightness of the light spot at various azimuth angles and the center is detected during OCT retinal imaging. The system automatically determines whether the light spot has completely entered the pupil area and adjusts the overall movement of the device and the reference arm based on the determination result to ensure clear retinal imaging. For example... Figure 4 As shown, when the area of ​​the light spot 15 at the fixed point is fully within the pupil 4, normal imaging will occur, and the scanning area 16 will not be obstructed; as Figure 5 As shown, when the light spot 15 at the fixed point does not completely enter the pupil 4, part of the scanning area 16 will be blocked, resulting in poor image quality.

[0084] The method provided in this embodiment will be used to achieve clear OCT retinal imaging.

[0085] For the acquired OCT retinal images, normalization is first performed, and then continuous sampling is performed. For each frame, median filtering is performed pixel-by-pixel to suppress impulse noise. Indicates the first indivual.

[0086] like Figure 6 As shown, taking a square scanning area 16 as an example, the scanning area 16 is divided into 5 regions, and the central region 17, the upper-left region 18 located at the upper left of the central region 17, the upper-right region 19 located at the upper right of the central region 17, the lower-left region 20 located at the lower left of the central region 17, and the lower-right region 21 located at the lower right of the central region 17 are shown. Specifically, in this embodiment:

[0087] Rc represents the central region 17, which occupies 20% × 20% of the image's width and height.

[0088] R1, R2, R3, and R4 represent the upper left region 18, the upper right region 19, the lower left region 20, and the lower right region 21, respectively, with each region occupying 15% × 15% of the image's width and height.

[0089] First, the brightness features of each region need to be extracted. In this embodiment, the five brightness feature values ​​will be calculated separately.

[0090] (1) Calculate the average brightness of each region, including the average brightness of the central region 17. and average brightness of the four corner areas :

[0091]

[0092]

[0093] in, The average brightness of the central area; The average brightness of the four corner areas. The numbers of the four corner sub-regions, The value can be 1, 2, 3, or 4, corresponding to the four sub-regions: top left, top right, bottom left, and bottom right, respectively. This indicates the average brightness of the upper left corner area. This indicates the average brightness of the upper right corner area. This indicates the average brightness of the lower left corner area. This indicates the average brightness of the lower right corner area; For the corresponding pixel brightness, Represents pixel coordinates, Represents the x-coordinate of pixels. Represents the ordinate of a pixel; Indicates the central area 17; Indicates the four corner sub-areas. This indicates the top-left sub-region. This indicates the upper right sub-region. This indicates the lower left sub-region. This indicates the lower right sub-region; This indicates calculating the average value.

[0094] (2) Calculate the global average brightness :

[0095]

[0096] in, This represents the preset minimum positive value, taken as... To prevent the denominator of the ratio from being 0.

[0097] This formula is a weighted average of the center brightness and the four corner brightness. This formula can highlight the weight of the central area 17, suppress extreme brightness interference, and more effectively reflect the overall brightness distribution of the image.

[0098] (3) Calculate the average brightness of the four corners :

[0099]

[0100] This formula represents the normalized average of the difference between the brightness at the four corners and the brightness at the center, through... Weighting the brightness difference between the center and corners (normalized term) can effectively suppress interference caused by excessive brightness differences between the corners and the center.

[0101] By including the brightness of the central area when calculating the average brightness of the four corners, it is possible to determine whether the light spot is evenly distributed within the pupil and to suppress local brightness anomalies in individual corners, thus avoiding misjudgment.

[0102] (4) Calculate the relative center ratio of the four corners :

[0103]

[0104] This formula represents the average difference between "corner brightness / center brightness" and "brightness difference normalization term", which can quantify the balance of brightness ratio between the four corners and the center. The closer to 1, the more consistent the brightness of the four corners with the center; a deviation from 1 indicates uneven brightness distribution.

[0105] Furthermore, based on the relationship between the brightness characteristics of the five regions and preset thresholds, the relative positions of the light spot and the pupil in the OCT retinal image are determined:

[0106] First, two types of thresholds are set. One is the absolute brightness threshold, which is the critical value that can distinguish between effective imaging and weak signals by statistically comparing the brightness of multiple sets of normal and abnormal OCT images. Secondly, based on the distribution of the brightness ratio of the four corners / center when the light spot fully enters the pupil and when it is off-center, the optimal brightness ratio threshold for classification is obtained. ;

[0107] By combining two indicators—whether the overall brightness is too low and the brightness ratio of the four corners relative to the center—three states, namely State A, State B, and State C, can be distinguished. The specific judgment rules are shown in Table 1.

[0108] Table 1. Rules for determining the relative position of the light spot and the pupil in OCT retinal images.

[0109]

[0110] In other words:

[0111] when > And all < When the light spot is concentrated near the edge of the pupil, it is marked as state A. At this time, the image shows a bright center but dark corners.

[0112] when > But some < When the light spot partially enters the pupil, it is marked as state B. At this time, the image shows a partially dark corner.

[0113] when > and > When the brightness of the center and four corners meets the requirements, it is marked as state C, at which point the image presents a globally bright state.

[0114] At this point, the image can be adjusted accordingly by combining the three states to determine the current position of the human eye relative to fixed point 6, and to judge the adjustment direction and accuracy. In this embodiment, a corresponding error function is constructed, and a PID control method is used to track the target brightness value to achieve clear presentation. Specifically:

[0115] (1) State A:

[0116] The center is bright but the corners are dark, indicating that the light spot is concentrated near the edge of the pupil. A PID algorithm is used to adjust the device's displacement, and through PID closed-loop control, the OCT image clarity is improved. maximize.

[0117] Based on the target brightness values ​​of each azimuth angle and average brightness of the four corners By constructing an error function model, we can obtain the calculation formula for the error function in PID control:

[0118]

[0119]

[0120] in, The error function in PID control measures the difference between the target sharpness and the actual sharpness. It is the core input of the PID controller, used to calculate the motor's displacement control signal. , Corresponding to z-axis movement This drives the device to adjust its position, maximizing the clarity of the OCT image. The target brightness value is determined by the average brightness of the central region when the light spot fully enters the pupil and the image is clear, and is used as the ideal target value for PID tracking. Indicates the current time Average brightness in the central region; Indicates the current time The average brightness of the four corners; , The brightness weights are for the center and the four corners, respectively.

[0121] until At this point, it is assumed that the light spot coincides with the pupil. However, the influence of deviations in the x and y directions of the light spot also needs to be considered.

[0122] It can be understood that the PID control model used in this embodiment is:

[0123]

[0124] in, The proportionality coefficient determines the immediate response of the error to the adjustment amount. These are the integral coefficients used to eliminate static errors and ensure that the final sharpness converges to the target value; These are differential coefficients, used to improve stability. This indicates that within the integration interval, the time... The changing error signal, i.e., in PID control, at any given time... Below, the deviation between the system's expected output and the actual output.

[0125] (2) State B:

[0126] Partially dark corners usually indicate that the light spot is off-center or tilted, requiring detection, judgment, and correction to ensure the light spot is aligned correctly.

[0127] If the light spot is off-center (offset relative to the center of the pupil), it will cause one or two corners on one side or opposite corners to be darker (obscured by the edge of the pupil).

[0128] If the light spot is deflected (the angle of incidence changes), it will cause one side or one opposite corner to remain dark (biased to one side).

[0129] At this point, first calculate the error of eccentricity or skewness:

[0130] X-direction error :

[0131]

[0132] like This indicates that the right side is brighter than the left side—the light spot is off to the left (the light needs to be moved to the right or the incident light needs to be shifted to the right).

[0133] Y-direction error :

[0134]

[0135] like If the upper side is brighter than the lower side, the light spot is biased downwards (the light needs to be moved upwards or the incident light needs to be biased upwards).

[0136] This formula, by normalizing and weighting the brightness of a single-sided region and the brightness difference within the combined region, can effectively distinguish between the overall spot offset and the interference of local brightness noise, and can also remove the influence of the absolute value of the overall brightness, directly outputting the relative offset degree in the X and Y directions.

[0137] At the same time, the result of this formula can directly correspond to the offset direction and can be accurately substituted into the error function for subsequent calculations.

[0138] If both left-right and up-down errors are observed to exhibit significant directional variation with small angle adjustments, then an angular skew is determined to exist.

[0139] After determining the state of the light spot eccentricity or skewness, the PID error function needs to be updated. ,join in , To make the adjustment more accurate. Final error function. The formula is:

[0140]

[0141]

[0142] in and For the newly added X-direction error and Y-direction error The weights are determined by substituting the error function into the PID control formula to make adjustments more efficient until the image is clearest.

[0143] (3) State C:

[0144] At this point, the image is globally bright, with the light spot fully entering the pupil. Using clinically accepted image sharpness metrics as a benchmark, a minimum threshold for meeting diagnostic requirements is determined. At this point, only fine-tuning of the reference arm's optical path is needed, while simultaneously observing whether the image sharpness exceeds the threshold. This is to determine if the image is at its sharpest. Once the image is clear, simply adjust the optical path of the reference arm to bring the retinal image to the appropriate acquisition position.

[0145] like Figure 7 The flowchart shown is the brightness state judgment and PID adjustment flowchart of this embodiment. This flowchart fully illustrates the entire process logic of the OCT retinal imaging adaptive adjustment method:

[0146] The process begins with system initialization, which first involves scanning the human eye from various angles to locate the light spot in the pupil area;

[0147] Simultaneously, a center + four-quadrant partitioning detection method is used to statistically analyze the brightness distribution;

[0148] Then, based on the brightness data, the state of the light spot is determined: one is "bright in the center but dark in the four corners" caused by the light spot being concentrated at the edge of the pupil, or "partially dark in the corners" caused by the light spot partially entering the pupil, or "globally bright" after the light spot has fully entered the pupil;

[0149] An error function is constructed for different states and substituted into the control formula to dynamically adjust the device displacement to optimize the spot entrance pupil range;

[0150] If the image does not reach the preset sharpness, the PID adjustment is executed cyclically until the image is clear. Then, the optical path of the reference arm is finely adjusted to place the image at the appropriate acquisition position.

[0151] The method provided in this embodiment solves the problems of incomplete light spot entry into the pupil and blurred imaging caused by pupil differences in traditional OCT imaging through partitioned brightness analysis and precise PID control strategy. Ultimately, it enables the device to adapt to different human eyes and achieve high-quality retinal imaging effect.

[0152] Example 2

[0153] This embodiment discloses an OCT retinal partition brightness adaptive clear imaging adjustment system.

[0154] The OCT retinal zone brightness adaptive clear imaging adjustment system includes:

[0155] The multi-region determination module is configured to: acquire OCT retinal images and determine five regions, including the center and four corners of the OCT retinal images;

[0156] The brightness feature extraction module is configured to extract brightness features of five regions, including the average brightness of the central region, the average brightness of the four corner regions, the global average brightness, the average brightness of the four corners, and the ratio of the four corners to the center.

[0157] The spot location module is configured to determine the relative position of the spot in the OCT retinal image and the pupil based on the relationship between the brightness characteristics of five regions and a preset threshold.

[0158] The PID tracking and adjustment module is configured to: construct a corresponding error function based on the relative position of the light spot and the pupil in the OCT retinal image, and use the PID control method to track the target brightness values ​​of five regions to adjust the position of the light spot so that the OCT retinal imaging achieves the preset clarity.

[0159] Example 3

[0160] The purpose of this embodiment is to provide a computer-readable storage medium.

[0161] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the OCT retinal partition brightness adaptive clear imaging adjustment method as described in Embodiment 1 of this disclosure.

[0162] Example 4

[0163] The purpose of this embodiment is to provide an electronic device.

[0164] An electronic device includes a memory, a processor, and a program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps in the OCT retinal partition brightness adaptive clear imaging adjustment method as described in Embodiment 1 of this disclosure.

[0165] The steps and methods involved in the apparatuses of Embodiments 2, 3, and 4 above correspond to those in Embodiment 1. For specific implementation details, please refer to the relevant description section of Embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood as including any medium capable of storing, encoding, or carrying an instruction set for execution by a processor and enabling the processor to perform any of the methods in this invention.

[0166] Those skilled in the art will understand that the modules or steps of the present invention described above can be implemented using general-purpose computer devices. Optionally, they can be implemented using computer-executable program code, thereby allowing them to be stored in a storage device for execution by a computer device, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. The present invention is not limited to any particular combination of hardware and software.

[0167] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. An OCT retinal zone brightness adaptive clear imaging adjustment method, characterized in that, Includes the following steps: Acquire an OCT retinal image and determine five regions in total, including the center and four corners of the OCT retinal image; The four corner regions refer to the upper left, upper right, lower left, and lower right corners of the OCT retinal image; The brightness features of five regions are extracted, including the average brightness of the central region, the average brightness of the four corner regions, the global average brightness, the average brightness of the four corners, and the ratio of the four corners to the center. Specifically, the average brightness of the central region is calculated by averaging the pixel brightness of the four corner regions; the average brightness of the four corner regions is calculated by averaging the pixel brightness of the four corner regions; the global average brightness is calculated by combining the weighted average of the average brightness of the central region and the average brightness of the four corner regions; a normalization term for the brightness difference between the center and the corners is set, and the average brightness of the four corners is calculated by weighting the normalization term for the brightness difference between the center and the corners; the ratio of the average brightness of the four corner regions to the average brightness of the central region is used, and the average difference between the normalization term for the brightness difference between the central region and the average brightness of the four corner regions is used to calculate the ratio of the four corners to the center. Based on the relationship between the brightness characteristics of five regions and preset thresholds, the relative positions of the light spots and pupils in the OCT retinal image are determined. Based on the relative position of the light spot and the pupil in the OCT retinal image, a corresponding error function is constructed. The PID control method is used to track the target brightness values ​​of five regions, thereby adjusting the position of the light spot and enabling the OCT retinal imaging to achieve the preset clarity.

2. The OCT retinal zone brightness adaptive clear imaging adjustment method as described in claim 1, characterized in that, The average brightness of the central area is calculated using the following formula: ； The average brightness of the four corner areas is calculated using the following formula: ； in, The average brightness of the central area; The average brightness of the four corner areas. The numbers of the four corner sub-regions, The value can be 1, 2, 3, or 4, corresponding to the four sub-regions: top left, top right, bottom left, and bottom right, respectively. This indicates the average brightness of the upper left corner area. This indicates the average brightness of the upper right corner area. This indicates the average brightness of the lower left corner area. This indicates the average brightness of the lower right corner area; For the corresponding pixel brightness, Represents pixel coordinates, Represents the x-coordinate of pixels. Represents the ordinate of a pixel; Indicates the central area; Indicates the four corner sub-areas. This indicates the top-left sub-region. This indicates the upper right sub-region. This indicates the lower left sub-region. This indicates the lower right sub-region; This indicates calculating the average value.

3. The OCT retinal zone brightness adaptive clear imaging adjustment method as described in claim 2, characterized in that, The global average brightness is calculated using the following formula: ； The formula for calculating the average brightness at the four corners is as follows: ； The ratio of the four corners to the center: ； in, Indicates the global average brightness; Indicates the average brightness of the four corners; Indicates the ratio of the four corners to the center; For a preset positive value, take To prevent the denominator of the ratio from being 0.

4. The OCT retinal zone brightness adaptive clear imaging adjustment method as described in claim 3, characterized in that, Based on the relationship between the brightness characteristics of five regions and preset thresholds, the relative position of the light spot in the OCT retinal image and the pupil is determined, specifically including: Two types of thresholds are preset: one is the absolute brightness threshold. Secondly, the optimal brightness ratio threshold for classification. ; when > And all < When the light spot is concentrated near the edge of the pupil, it is marked as state A; when > But some < When the light spot partially enters the pupil, it is marked as state B; when > and > When the brightness of the center and four corners meets the requirements, it is marked as state C.

5. The OCT retinal zone brightness adaptive clear imaging adjustment method as described in claim 4, characterized in that, When in state A, the OCT retinal image appears bright in the center but dark in the four corners. Based on the target brightness value, the average brightness of the central region, and the average brightness of the four corners, an error function is constructed: ； ； in, This is the error function in PID control; The target brightness value; Indicates the current time Average brightness in the central region; Indicates the current time The average brightness of the four corners; Indicates time; and The brightness weights are for the center and the four corners, respectively.

6. The OCT retinal zone brightness adaptive clear imaging adjustment method as described in claim 4, characterized in that, When in state B, the OCT retinal image shows partial angular darkness, indicating that the light spot is off-center or tilted. In this case, the error of off-center or tilt is first calculated, and then an error function is constructed based on the error of off-center or tilt: ； ； in, This is the error function in PID control; The target brightness value; Indicates the current time Average brightness in the central region; Indicates the current time The average brightness of the four corners; Indicates time; and These represent the errors in the X and Y directions, respectively. , , and These represent the brightness weights for the center, four corners, X-direction error, and Y-direction error, respectively.

7. The OCT retinal zone brightness adaptive clear imaging adjustment method as described in claim 6, characterized in that, The specific formulas for calculating the X-direction error and the Y-direction error are as follows: ； 。 8. The OCT retinal zone brightness adaptive clear imaging adjustment method as described in claim 6, characterized in that: like This indicates that the right side is brighter than the left side, and the light spot is biased to the left. In this case, it is necessary to move the light to the right or shift the incident light to the right. like This indicates that the left side is brighter than the right side, and the light spot is biased to the right. In this case, it is necessary to move the light to the left or shift the incident light to the left. like This indicates that the upper side is brighter than the lower side, and the light spot is lower. In this case, it is necessary to move the light upward or shift the incident light upward. like This indicates that the lower side is brighter than the upper side, and the light spot is biased upwards. In this case, it is necessary to move the light downwards or shift the incident light downwards.

9. The OCT retinal zone brightness adaptive clear imaging adjustment method as described in claim 1, characterized in that, It also includes a method that if the target brightness value is not reached after tracking it using the constructed error function, the PID adjustment is executed cyclically until the image reaches the preset sharpness.

10. An OCT retinal zone brightness adaptive clear imaging adjustment system, characterized in that, include: The multi-region determination module is configured to: acquire OCT retinal images and determine five regions, including the center and four corners of the OCT retinal images; The four corner regions refer to the upper left, upper right, lower left, and lower right corners of the OCT retinal image; The brightness feature extraction module is configured to extract brightness features of five regions, including the average brightness of the central region, the average brightness of the four corner regions, the global average brightness, the average brightness of the four corners, and the ratio of the four corners to the center. Specifically, the average brightness of the central region is calculated by averaging the pixel brightness of the four corner regions; the average brightness of the four corner regions is calculated by averaging the pixel brightness of the four corner regions; the global average brightness is calculated by combining the weighted average of the average brightness of the central region and the average brightness of the four corner regions; a normalization term for the brightness difference between the center and the corners is set, and the average brightness of the four corners is calculated by weighting the normalization term for the brightness difference between the center and the corners; the ratio of the average brightness of the four corner regions to the average brightness of the central region is used, and the average difference between the normalization term for the brightness difference between the central region and the average brightness of the four corner regions is used to calculate the ratio of the four corners to the center. The spot location module is configured to determine the relative position of the spot in the OCT retinal image and the pupil based on the relationship between the brightness characteristics of five regions and a preset threshold. The PID tracking and adjustment module is configured to: construct a corresponding error function based on the relative position of the light spot and the pupil in the OCT retinal image, and use the PID control method to track the target brightness values ​​of five regions to adjust the position of the light spot so that the OCT retinal imaging achieves the preset clarity.

11. A computer-readable storage medium having a program stored thereon, characterized in that, When executed by the processor, the program implements the steps in the OCT retinal partition brightness adaptive clear imaging adjustment method as described in any one of claims 1-9.

12. An electronic device comprising a memory, a processor, and a program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the OCT retinal partition brightness adaptive clear imaging adjustment method as described in any one of claims 1-9.

Images