Laser fundus camera shooting method and electronic equipment

Abstract

The invention relates to a laser fundus camera shooting method and electronic equipment, and belongs to the technical field of medical image.The method comprises the steps that a fundus near-infrared image collected by a laser fundus camera in real time is obtained; extracting a light spot region in the fundus near-infrared image, and calculating position feature data and first area proportion data of the light spot region and the fundus near-infrared image; adjusting the spatial position of the laser fundus camera on the basis of the position feature data and the area proportion data, so that the first area proportion data is greater than a preset first threshold value, and completing alignment; and after alignment is completed, focusing adjustment is carried out, and after focusing is completed, the laser fundus camera is controlled to execute a shooting task. According to the method, a pupil camera does not need to be used, the retina area can be automatically aligned and tracked only through the fundus near-infrared image, the cost is effectively reduced, meanwhile, the position deviation of the pupil camera and the laser fundus camera does not need to be considered, and the alignment precision is effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical imaging, in particular to a laser fundus camera shooting method and an electronic device. BACKGROUND

[0002] The laser fundus camera is a key tool for screening, diagnosis and follow-up of ophthalmic diseases. It can obtain high-contrast retinal images by emitting laser of specific wavelength to the fundus and receiving reflected light, which is used for detecting eye diseases. High-quality fundus images are the basis for accurate diagnosis, and the traditional process of obtaining such images relies heavily on the professional skills of the operator. The operator needs to manually control the device to complete the three core steps of accurate alignment of the patient's pupil (pupil alignment), adjustment of the focal length to make the retina clearly imaged (focusing), and adjustment of the laser intensity to make the image brightness appropriate (brightness adjustment) in turn. This process not only takes time, but also requires high experience of the operator, resulting in low efficiency and difficulty in ensuring the consistency of the results, which limits the popularization of the technology in primary medical care or large-scale screening scenarios.

[0003] To reduce the operation difficulty and improve the efficiency, the existing technology has developed an automatic shooting scheme. This scheme usually relies on an additional pupil camera (or external eye camera). Its workflow is as follows: first, the pupil camera captures the external eye image in real time, and the system identifies the pupil center position through image processing algorithm, calculates the adjustment amount required for the three-axis motor (controls the laser fundus camera to move in left-right, up-down and front-back directions) according to the deviation of the pupil center from the image center, and drives the laser fundus camera lens to align with the pupil. After completing the pupil alignment, the system switches to the analysis based on the fundus image to perform automatic focusing and brightness adjustment, and finally triggers the shooting.

[0004] However, the above scheme has the following defects: first, an additional pupil camera is needed, and factors such as its volume, installation position, power supply and data transmission also need to be considered, thereby increasing the system hardware cost; second, the core target of alignment is to align the optical axis of the laser fundus camera with the center of the retina, and there is a positional deviation between the optical axes of the pupil camera and the laser fundus camera, so using the pupil camera to assist the laser fundus camera to align with the center of the retina is easy to introduce secondary deviation, thereby reducing the alignment accuracy of the laser fundus camera. SUMMARY

[0005] In view of this, it is necessary to provide a laser fundus camera shooting method and an electronic device to solve the technical problems of high system hardware cost and low accuracy of the laser fundus camera in the prior art.

[0006] To solve the above problems, on the one hand, the present application provides a laser fundus camera shooting method, comprising: obtaining a fundus near-infrared image collected by a laser fundus camera in real time; Extract the spot region from the fundus near-infrared image, and calculate the positional feature data and first area ratio data of the spot region relative to the fundus near-infrared image; The spatial position of the laser fundus camera is adjusted based on the location feature data and the area ratio data so that the first area ratio data is greater than a preset first threshold, thereby completing the alignment. After alignment is completed, focus adjustment is performed, and once focus is achieved, the laser fundus camera is controlled to perform the shooting task.

[0007] In one possible implementation, extracting the spot region from the near-infrared fundus image includes: The near-infrared image of the fundus is binarized to obtain an initial binary image; The initial binary image is subjected to morphological processing to obtain the target binary image; Connectivity analysis is performed on the target binary image to extract the area of ​​the largest light spot.

[0008] In one possible implementation, adjusting the spatial position of the laser fundus camera based on the location feature data and the first area proportion data includes: Based on the positional feature data, a first adjustment amount and a second adjustment amount of the laser fundus camera in a first direction and a second direction in a second direction are calculated, wherein the first direction and the second direction are perpendicular. Based on the first area ratio data, the third adjustment amount of the laser fundus camera in the third direction is calculated, wherein the third direction is perpendicular to both the first direction and the second direction; The spatial position of the laser fundus camera is adjusted based on the first adjustment amount, the second adjustment amount, and the third adjustment amount.

[0009] In one possible implementation, the method further includes, before alignment is completed: Calculate the first average gray value of the near-infrared fundus image; The laser intensity of the laser fundus camera is adjusted based on the first average gray value so that the first average gray value is within a preset first gray value range.

[0010] In one possible implementation, focus adjustment includes: Extract the central region of the fundus near-infrared image, and calculate the sharpness of the central region based on the grayscale value of the central region; The optimal sharpness is extracted from the sharpness at different image distances of the laser fundus camera, and the image distance of the laser fundus camera is adjusted to the image distance corresponding to the optimal sharpness.

[0011] In one possible implementation, extracting the central region of the near-infrared fundus image includes: Using the center of the near-infrared fundus image as the center and a preset value as the radius, a circular region is extracted; Extract the central region from the circular region whose grayscale value is higher than the preset grayscale threshold.

[0012] In one possible implementation, extracting the optimal sharpness from the sharpness at different image distances of the laser fundus camera, and adjusting the image distance of the laser fundus camera to the image distance corresponding to the optimal sharpness, includes: The focusing lens group of the laser fundus camera is controlled to move within its range of motion at a preset first speed to obtain the clarity of the focusing lens group at different positions within the range of motion. Extract candidate locations corresponding to peak values ​​from the sharpness at different locations within the movement range, and determine the candidate range based on the candidate locations; The focusing lens group of the laser fundus camera is controlled to move at a preset second speed within the candidate range to obtain the clarity of the focusing lens group at different positions within the candidate range, wherein the first speed is greater than the second speed; The optimal position corresponding to the best sharpness is extracted from the sharpness at different positions within the candidate range, and the focusing lens group is controlled to move to the optimal position.

[0013] In one possible implementation, the method further includes: Calculate the second average gray value of the central region; The laser intensity of the laser fundus camera is adjusted based on the second average gray value so that the second average gray value is within a preset second gray value range.

[0014] In one possible implementation, the method further includes: Calculate the second area ratio data between the central region and the near-infrared image of the fundus; When the second area percentage data is consistently no greater than the preset second threshold, return to the alignment step.

[0015] On the other hand, this application also provides an electronic device, including a memory and a processor; The memory is used to store programs; The processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps of the laser fundus camera imaging method described above.

[0016] The beneficial effects of this application are: the laser fundus camera imaging method provided by this application does not require the use of a pupil camera, and can automatically align and track the retinal area using only near-infrared images of the fundus. The hardware system design is lighter and simpler, effectively reducing costs. At the same time, it does not require consideration of the positional deviation between the pupil camera and the laser fundus camera, effectively improving alignment accuracy.

[0017] Furthermore, this application allows for simultaneous adjustment of the focal length and laser intensity during the alignment process, enabling fine-tuning during focusing to quickly achieve the optimal image capture state. This achieves the simultaneous solution of spatial alignment and image quality optimization using near-infrared fundus images, creating a deeply intertwined and mutually reinforcing integrated process for alignment, focusing, and brightening, effectively improving overall efficiency. Attached Figure Description

[0018] Figure 1 A schematic flowchart of an embodiment of the laser fundus camera imaging method provided in this application; Figure 2 This is a schematic flowchart of an embodiment of step S102 of this application; Figure 3 This is a schematic flowchart of an embodiment of step S103 of this application; Figure 4 This is a schematic flowchart of an embodiment of step S104 of this application; Figure 5 This is a schematic flowchart of an embodiment of step S401 of this application; Figure 6 This is a schematic flowchart of an embodiment of step S402 of this application; Figure 7 This is a schematic flowchart of an embodiment of step S4025 of this application; Figure 8 A schematic flowchart of an embodiment of the brightness adjustment steps provided in this application; Figure 9 A schematic flowchart of an embodiment of the reset steps provided in this application; Figure 10 A schematic diagram of an embodiment of the electronic device provided in this application. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0020] It should be understood that the illustrative drawings are not drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of this application. It should be understood that the operations in the flowcharts may be implemented out of order, and steps without logical contextual relationships may be reversed or performed simultaneously. Furthermore, those skilled in the art, guided by the content of this application, may add one or more other operations to the flowcharts, or remove one or more operations from the flowcharts. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor systems and / or microcontroller systems.

[0021] The terms "first," "second," etc., used in the embodiments of this application are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a technical feature specified with "first" or "second" may explicitly or implicitly include at least one of those features. "And / or" describes the relationship between related objects, indicating that three relationships may exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone.

[0022] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0023] This application provides a laser fundus camera imaging method and electronic device, which are described below.

[0024] Figure 1 This is a schematic flowchart of an embodiment of the laser fundus camera imaging method provided in this application, as shown below. Figure 1 As shown, the laser fundus camera imaging method includes: S101. Acquire near-infrared images of the fundus captured in real time by a laser fundus camera; It should be noted that the near-infrared (IR) images of the fundus captured by the laser fundus camera are formed by the reflection of light emitted from a near-infrared laser source onto the retinal surface after passing through the pupil. In actual use, the subject's chin is stabilized by a fixation device, resulting in an initial distance between the camera lens and the eyeball, and they are not yet precisely aligned. The incident light path deviates from the optical axis of the eyeball, causing the camera to only capture reflection information from a localized area of ​​the retina. The resulting near-infrared fundus images typically appear as small bright spots or low-contrast, small-field-of-view images, with the overall effective imaging range relatively limited, accounting for approximately [missing information - likely a percentage of the total area]. Figure 10 %~20%.

[0025] S102. Extract the spot region from the near-infrared image of the fundus, and calculate the positional feature data and the first area ratio data of the spot region relative to the near-infrared image of the fundus. In practical applications, ideally only the light spot area should appear as a continuous bright area. However, in reality, the light spot area is often not completely continuous, containing dark areas, holes, or breaks, and surrounded by discrete bright areas. These phenomena are mainly caused by corneal reflection, blood vessels or eyelashes obstructing the view, and low contrast. To solve the above problems and improve the accuracy of image extraction, in some embodiments of this application, such as... Figure 2 As shown, step S102 involves extracting the spot region from the near-infrared image of the fundus, including: S1021. Binarize the near-infrared image of the fundus to obtain an initial binary image; It should be noted that binarization can be based on threshold segmentation, where bright pixels with gray values ​​greater than the threshold are set to 1, and dark pixels with gray values ​​less than the threshold are set to 0, forming an initial binary image. The threshold can be a fixed threshold, an adaptive threshold, a dual threshold, etc., and is not limited here. In addition, the fundus near-infrared image is usually smoothed by Gaussian (5×5) before binarization to suppress noise.

[0026] S1022. Perform morphological processing on the initial binary image to obtain the target binary image; It should be noted that the purpose of morphological processing is to fill the disconnected areas of the small hole connection. Specifically, it can be a morphological closing operation, which first expands and connects the adjacent bright pixel areas, and then erodes to restore the approximate shape. It can also be an opening operation, etc., which are not limited here.

[0027] S1023. Perform connected component analysis on the target binary image to extract the spot region with the largest area.

[0028] It should be noted that the purpose of extracting the largest spot area is to filter out interfering bright areas.

[0029] Based on the extracted spot area, the positional feature data and the first area ratio data of the spot area and the fundus near-infrared image are calculated. Specifically, the positional feature data is the deviation between the centroid coordinates of the spot area and the center coordinates of the fundus near-infrared image, and the first area ratio data is the ratio of the area of ​​the spot area to the area of ​​the entire fundus near-infrared image.

[0030] S103. Adjust the spatial position of the laser fundus camera based on the position feature data and the first area proportion data so that the first area proportion data is greater than the preset first threshold, and complete the alignment. In some embodiments of this application, such as Figure 3 As shown, step S103, which adjusts the spatial position of the laser fundus camera based on location feature data and the first area proportion data, includes: S1031. Based on positional feature data, calculate the first adjustment amount of the laser fundus camera in the first direction and the second adjustment amount in the second direction, wherein the first direction and the second direction are perpendicular. It should be noted that the first direction is the X-axis parallel to the coronal axis (left and right), and the first adjustment amount in the first direction is (X1-Xc) / r×P, where X1 is the x-axis coordinate of the centroid of the current frame spot region, Xc is the x-axis coordinate of the center of the fundus near-infrared image, r is the unit spatial resolution in pixels / μm, and P is the conversion factor between the fundus physical distance and the number of motor pulses in pulses / μm; the second direction is the Y-axis parallel to the vertical axis (up and down), and the second adjustment amount in the second direction is (Y1-Yc) / r×P, where Y1 is the y-axis coordinate of the centroid of the current frame spot region, and Yc is the y-coordinate of the center of the fundus near-infrared image.

[0031] S1032. Based on the first area ratio data, calculate the third adjustment amount of the laser fundus camera in the third direction, where the third direction is perpendicular to both the first and second directions; It should be noted that the third direction is specifically the Z-axis parallel to the sagittal axis (closer to / farther from the eyeball); the third adjustment amount is Ac / Area×Pa, where Ac is the pixel area of ​​the spot region in the current frame, Area is the total area of ​​the near-infrared image of the fundus, and Pa is the conversion factor between the area ratio and the number of motor pulses.

[0032] S1033. Adjust the spatial position of the laser fundus camera based on the first adjustment amount, the second adjustment amount, and the third adjustment amount.

[0033] Furthermore, in order to provide better brightness conditions for the alignment step and better initial brightness conditions for the subsequent focusing step, and to ensure the normal operation of the image processing algorithm, in some embodiments of this application, before completing the alignment, the method further includes: calculating a first average gray value of the near-infrared image of the fundus, and adjusting the laser intensity of the laser fundus camera based on the first average gray value so that the first average gray value is within a preset first gray range.

[0034] Based on the calculated three-axis adjustment values ​​(i.e., the first adjustment value, the second adjustment value, and the third adjustment value) and the current position of the three-axis motor, the target position of the three-axis motor can be obtained. The three-axis motor moves to the target position according to the movement direction and speed commands issued by the algorithm. During the movement, the near-infrared image of the fundus is continuously collected to update the movement commands of the three-axis motor to form a closed loop. When the first area ratio data is greater than the preset first threshold, that is, when the near-infrared image of the fundus is in a stable and full state, it is determined that the optical axis of the laser fundus camera is aligned with the center of the retina and the first average gray value is within the preset first gray value range. The alignment is considered to be complete, and then the autofocus process is started.

[0035] It should also be noted that during the autofocus process, the alignment enters the follow mode, which means that the eye movement offset is calculated based on the deviation between the centroid of the spot area and the center of the near-infrared image of the fundus and the follow mode. Specifically, the first adjustment amount and the second adjustment amount are calculated in real time. The spatial position of the laser fundus camera in the first direction is adjusted based on the first adjustment amount, and the spatial position of the laser fundus camera in the second direction is adjusted based on the second adjustment amount.

[0036] S104. After alignment, adjust the focus and control the laser fundus camera to perform the shooting task.

[0037] Considering issues such as eyelashes, eyelids obstructing focus, and edge interference, in some embodiments of this application, only the sharpness of the central area is calculated for focus adjustment, such as... Figure 4 As shown, focusing in step S104 includes: S401. Extract the central region of the near-infrared image of the fundus and calculate the clarity of the central region based on the gray value of the central region. To better extract the central region, in some embodiments of this application, such as Figure 5 As shown, in step S401, extracting the central region of the near-infrared image of the fundus includes: S4011. Using the center of the near-infrared image of the fundus as the center and the preset value as the radius, extract a circular region; S4012. Extract the central region from the circular region whose grayscale value is higher than the preset grayscale threshold.

[0038] Based on the extracted central region, the sharpness is calculated using the formula: FBrenner=∑M∑N(f(x+2,y)-f(x,y)) 2 Where M is the width of the central region, N is the height of the central region, f(x+2,y) is the gray value at pixel coordinates (x+2,y) in the image, and f(x,y) is the gray value at pixel coordinates (x,y) in the image. By iterating through each pixel in the central region, the square of the gray value of the point is calculated and the gray value of the point is calculated to be two pixels away from it in the horizontal direction. The calculation results of all pixels are accumulated to obtain the sharpness.

[0039] S402. Extract the optimal sharpness from the sharpness at different image distances of the laser fundus camera, and adjust the image distance of the laser fundus camera to the image distance corresponding to the optimal sharpness.

[0040] In some embodiments of this application, such as Figure 6 As shown, step S402 specifically includes: S4021. Control the focusing lens group of the laser fundus camera to move within its moving range at a preset first speed, and obtain the clarity of the focusing lens group at different positions within the moving range. It should be noted that this step corresponds to coarse adjustment, which specifically includes: first, controlling the focusing motor to drive the focusing lens group to move at a preset first speed and first step length from the starting point of the movement range to one end at a constant speed. During the movement, the laser fundus camera simultaneously scans and acquires near-infrared images of the fundus and extracts the central region to calculate the sharpness. When the peak of sharpness is found at one boundary of the movement range of the focusing lens group or at different positions on one side, the direction is immediately reversed. After scanning the other side, the curves at both ends are merged to obtain the sharpness of the focusing lens group at different positions within the movement range. The candidate position corresponding to the peak value is selected, and this candidate position is used as the optimal focus for coarse adjustment.

[0041] S4022. Extract candidate locations corresponding to peak values ​​from the sharpness at different locations within the movement range, and determine the candidate range based on the candidate locations; It should be noted that the candidate range is obtained by expanding a preset distance range to both sides of the candidate position.

[0042] S4023. Control the focusing lens group of the laser fundus camera to move within the candidate range at a preset second speed to obtain the clarity of the focusing lens group at different positions within the candidate range, wherein the first speed is greater than the second speed. It should be noted that this step corresponds to fine-tuning, and the second speed of fine-tuning is less than the first speed of coarse-tuning.

[0043] S4024. Extract the best position corresponding to the best sharpness from the sharpness at different positions within the candidate range; S4025, Control the focusing lens group to move to the optimal position.

[0044] Furthermore, to address the physiological micro-movements of the examinee's eyeballs, in some embodiments of this application, such as... Figure 7 As shown, step S4025 specifically includes: S701. When the error between the current position and the optimal position of the focusing lens group is less than the preset error threshold, obtain the variance of the sharpness within the preset time window. S702. When the variance of sharpness is less than the preset first variance threshold, control the focusing lens group to stop moving. S703. When the variance of sharpness is not less than the preset first variance threshold, the target step size and target movement direction of the focusing lens group are determined according to the step size, movement direction and sharpness change of the focusing lens group within the preset time window, and the movement of the focusing lens group is controlled based on the target step size and target movement direction.

[0045] Furthermore, in order to provide better brightness conditions for the focusing step, in some embodiments of this application, such as Figure 8 As shown, the method also includes: S801, Calculate the second average gray value of the central region; S802. Adjust the laser intensity of the laser fundus camera based on the second average gray value so that the second average gray value is within a preset second gray value range.

[0046] It should be noted that the second grayscale range corresponding to the focusing step is within the first grayscale range corresponding to the alignment step. This is because the goal of the alignment step is to quickly and stably locate the light spot and calculate its position and size, requiring a clear distinction between the light spot and the background. An image with moderate brightness and suitable contrast is suitable for stable thresholding and morphological processing, while avoiding overexposure leading to outline expansion of the light spot area or underexposure leading to outline breakage. Furthermore, during the alignment stage, the laser fundus camera moves rapidly, and the relative distance and angle with the eyeball change drastically, potentially causing a significant instantaneous change in the intensity of reflected light. Stabilizing the first average grayscale value at a relatively low and more conservative level allows for sufficient upper limits of dynamic range, preventing instantaneous overexposure due to sudden proximity or other factors. On the other hand, the goal of the focusing step is to finely evaluate the sharpness of image details in the aligned area, requiring the image to have rich texture details. Therefore, higher brightness is needed, hence the different grayscale ranges corresponding to the alignment and focusing steps.

[0047] Furthermore, to address situations where significant head movements or blinking by the examinee make it impossible to locate the central area, in some embodiments of this application, such as... Figure 9 As shown, the method also includes the following during the focusing process: S901, Calculate the second area ratio data of the central region and the fundus near-infrared image; S902. When the second area percentage data is consistently no greater than the preset second threshold, return to the alignment step.

[0048] It should be noted that, specifically, if the second area ratio data is consistently not greater than the preset second threshold, autofocus will stop, return to position mode, and reset to wait for re-alignment. The "consistently not greater than" condition is to prevent shake. In addition, if the sharpness peak does not increase for a long time and the sharpness or grayscale fluctuates greatly, it will return to coarse adjustment and reposition.

[0049] Considering the case of focusing alone, separate start and finish conditions are set for the focusing stage. The start conditions specifically include: the second area ratio data is greater than the preset second threshold, and the second average gray value is within the preset first gray range. The finish conditions specifically include: the second area ratio data is consistently greater than the preset second threshold, the optimal sharpness is determined, and the second average gray value is consistently within the preset second gray range.

[0050] The above method has two main application scenarios in practice: Fully automatic binocular imaging process: 1) After clicking the fully automatic function button for both eyes, it will start with any eye of the subject and automatically align (retina) + automatically adjust focus and brightness. After adjustment, it will automatically take fundus photos such as laser color / ICGA / FFA / IR. After taking photos of the first eye, it will automatically switch to the other eye and repeat the automatic pupil alignment + automatic focus and brightness adjustment process. After completion, it will automatically take photos of the second eye.

[0051] Semi-automatic shooting process for a single eye: 2) After clicking the single-eye semi-automatic function button, the device will determine the current left / right eye (OD / OS) partition position and automatically perform pupil alignment and automatic focus and brightness adjustment from the current eye. After adjustment, the doctor / technician can press the shooting button to acquire a high-quality fundus image. In addition, if the quality of the acquired image is not good, the automatic alignment, focus and brightness adjustment results will be manually fine-tuned before shooting to correct the issue.

[0052] Compared with existing technologies, this application does not require the use of a pupil camera. It can automatically align and track the retinal area using only near-infrared images of the fundus. The hardware system design is lighter and simpler, effectively reducing costs. At the same time, it does not need to consider the positional deviation between the pupil camera and the laser fundus camera, effectively improving alignment accuracy.

[0053] Furthermore, this application allows for simultaneous adjustment of the focal length and laser intensity during the alignment process, enabling fine-tuning during focusing to quickly achieve the optimal image capture state. This achieves the simultaneous solution of spatial alignment and image quality optimization using near-infrared fundus images, creating a deeply intertwined and mutually reinforcing integrated process for alignment, focusing, and brightening, effectively improving overall efficiency.

[0054] like Figure 10 As shown, this application also provides an electronic device. This electronic device includes at least a processor 1001 and a memory 1002.

[0055] Processor 1001 may include one or more processing cores, such as a quad-core processor, an octa-core processor, etc. Processor 1001 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). Processor 1001 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, processor 1001 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, processor 1001 may also include an AI (Artificial Intelligence) processor, which is used to handle computational operations related to machine learning.

[0056] The memory 1002 may include one or more computer-readable storage media, which may be non-transitory. The memory 1002 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory 1002 are used to store at least one instruction, which is executed by the processor 1001 to implement the laser fundus camera imaging method provided in the method embodiments of this application.

[0057] In some embodiments, the electronic device may also optionally include a peripheral device interface and at least one peripheral device. The processor 1001, memory 1002, and peripheral device interface can be connected via a bus or signal line. Each peripheral device can be connected to the peripheral device interface via a bus, signal line, or circuit board. Indicatively, peripheral devices include, but are not limited to, radio frequency circuits, touch displays, audio circuits, and power supplies.

[0058] Of course, electronic devices may also include fewer or more components, and this embodiment does not limit this.

[0059] The above provides a detailed description of a laser fundus camera imaging method provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

[0060] The above description is merely a preferred embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for capturing images with a laser fundus camera, characterized in that, include: Acquire near-infrared images of the fundus in real time using a laser fundus camera; Extract the spot region from the fundus near-infrared image, and calculate the positional feature data and first area ratio data of the spot region relative to the fundus near-infrared image; The spatial position of the laser fundus camera is adjusted based on the location feature data and the area ratio data so that the first area ratio data is greater than a preset first threshold, thereby completing the alignment. After alignment is completed, focus adjustment is performed, and once focus is achieved, the laser fundus camera is controlled to perform the shooting task.

2. The laser fundus camera imaging method according to claim 1, characterized in that, Extracting the spot region from the near-infrared image of the fundus includes: The near-infrared image of the fundus is binarized to obtain an initial binary image; The initial binary image is subjected to morphological processing to obtain the target binary image; Connectivity analysis is performed on the target binary image to extract the area of ​​the largest light spot.

3. The laser fundus camera imaging method according to claim 1, characterized in that, Adjusting the spatial position of the laser fundus camera based on the location feature data and the first area proportion data includes: Based on the positional feature data, a first adjustment amount and a second adjustment amount of the laser fundus camera in a first direction and a second direction in a second direction are calculated, wherein the first direction and the second direction are perpendicular. Based on the first area ratio data, the third adjustment amount of the laser fundus camera in the third direction is calculated, wherein the third direction is perpendicular to both the first direction and the second direction; The spatial position of the laser fundus camera is adjusted based on the first adjustment amount, the second adjustment amount, and the third adjustment amount.

4. The laser fundus camera imaging method according to claim 1, characterized in that, Before alignment is completed, the method further includes: Calculate the first average gray value of the near-infrared fundus image; The laser intensity of the laser fundus camera is adjusted based on the first average gray value so that the first average gray value is within a preset first gray value range.

5. The laser fundus camera imaging method according to claim 1, characterized in that, Focus adjustment, including: Extract the central region of the fundus near-infrared image, and calculate the sharpness of the central region based on the grayscale value of the central region; The optimal sharpness is extracted from the sharpness at different image distances of the laser fundus camera, and the image distance of the laser fundus camera is adjusted to the image distance corresponding to the optimal sharpness.

6. The laser fundus camera imaging method according to claim 5, characterized in that, Extracting the central region of the near-infrared fundus image includes: Using the center of the near-infrared fundus image as the center and a preset value as the radius, a circular region is extracted; Extract the central region from the circular region whose grayscale value is higher than the preset grayscale threshold.

7. The laser fundus camera imaging method according to claim 5, characterized in that, Extracting the optimal sharpness from the sharpness at different image distances of the laser fundus camera, and adjusting the image distance of the laser fundus camera to the image distance corresponding to the optimal sharpness, includes: The focusing lens group of the laser fundus camera is controlled to move within its range of motion at a preset first speed to obtain the clarity of the focusing lens group at different positions within the range of motion. Extract candidate locations corresponding to peak values ​​from the sharpness at different locations within the movement range, and determine the candidate range based on the candidate locations; The focusing lens group of the laser fundus camera is controlled to move at a preset second speed within the candidate range to obtain the clarity of the focusing lens group at different positions within the candidate range, wherein the first speed is greater than the second speed; The optimal position corresponding to the best sharpness is extracted from the sharpness at different positions within the candidate range, and the focusing lens group is controlled to move to the optimal position.

8. The laser fundus camera imaging method according to claim 5, characterized in that, The method further includes: Calculate the second average gray value of the central region; The laser intensity of the laser fundus camera is adjusted based on the second average gray value so that the second average gray value is within a preset second gray value range.

9. The laser fundus camera imaging method according to claim 5, characterized in that, The method further includes: Calculate the second area ratio data between the central region and the near-infrared image of the fundus; When the second area percentage data is consistently no greater than the preset second threshold, return to the alignment step.

10. An electronic device, characterized in that, Including memory and processor; The memory is used to store programs; The processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps of the laser fundus camera imaging method according to any one of claims 1 to 9.

Images