High-precision eyeball movement real-time detection system
Through high-definition high-speed camera module and animation video guidance technology, high-precision real-time detection of eye movements is achieved, error problems caused by indirect measurements in the existing technology are solved, and more accurate eye movement tracking and diagnostic reference is provided.
Patent Information
- Application Number
- CN202411137525.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-19
- Publication Date
- 2024-11-05
AI Technical Summary
The existing eye movement tracking technology is susceptible to interference due to indirect measurements and generates systematic errors, so it is impossible to track eye position with high accuracy in three-dimensional space.
The high-definition high-speed camera module is used to collect user eye image data, guide the eyeball to rotate through animated video, perform image segmentation, reconstruction and differential evaluation, and build a plane view of the user eyeball to achieve high-precision real-time detection of eye movement.
It provides more accurate eye movement detection, which can accurately track eye position in three-dimensional space, providing more reliable auxiliary reference data for the diagnosis of eye diseases.
Smart Images

Figure CN118892301A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of visual detection, and in particular to a high-precision real-time detection system for eye movement. Background Art
[0002] Eye movement is closely related to eye diseases. For example, nystagmus may be a manifestation of certain neurological diseases or intraocular diseases. Patients with strabismus have uncoordinated eye movements, and both eyes cannot focus on the same target at the same time. Paralytic strabismus can cause restricted eye movement in a specific direction. These abnormal eye movements are often important clues for the diagnosis of eye diseases.
[0003] The invention patent with application number 202110990930.6 discloses a contact lens eye movement tracking method, which is characterized in that the method includes: the MEMS acceleration sensor chip detects the acceleration data of the eyeball, and transmits the acceleration data to the main control unit; the main control unit detects the acceleration data to wake up the MEMS angular velocity sensor chip and the Hall magnetometer chip; the MEMS angular velocity sensor chip obtains the angular velocity data of the eyeball movement, and transmits the angular velocity data to the main control unit; the Hall magnetometer chip obtains geomagnetic data, and transmits the geomagnetic data to the main control unit at predetermined intervals; the main control unit corrects the angular velocity data according to the geomagnetic data, obtains the corrected angular velocity data and records it as target angular velocity data, and the main control unit calculates the eyeball movement trajectory according to the acceleration data, target angular velocity data and geomagnetic data, and transmits the eyeball movement trajectory to the mobile terminal.
[0004] The application aims to solve the problem that "eye tracking technology is easily interfered with and produces systematic errors, which limits its accuracy because it measures indirectly through optical or electrical signals rather than directly. It can only track the position of the eye relative to the head, but cannot track the position of the eye in three-dimensional free space."
[0005] However, there is currently no high-precision machine vision-based eye movement detection technology;
[0006] To this end, we propose a high-precision real-time eye movement detection system. Summary of the invention
[0007] In view of the above-mentioned shortcomings of the prior art, the present invention provides a high-precision real-time eye movement detection system, which solves the technical problems raised in the above-mentioned background technology.
[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions:
[0009] A high-precision real-time eye movement detection system, comprising:
[0010] A camera module is used to collect user eye image data; a receiving module is used to receive the user eye image data collected by the camera module and store the user eye image data; a segmentation module is used to obtain the user eye image data stored in the receiving module, perform segmentation processing on the user eye image data to obtain the user eye image and feed it back to the receiving module for storage; a reconstruction module is used to receive the user eye image and construct a user eye plane view based on the user eye image; a storage module is used to receive the user eye plane image constructed in the reconstruction module, construct a timestamp mark on the user eye plane image and store it; an evaluation module is used to receive the user eye plane image stored in the storage module and evaluate the difference of the user eye plane image.
[0011] Furthermore, the camera module is internally provided with submodules, including:
[0012] A guiding unit, used to play animation and provide a target for users to focus on;
[0013] Among them, during the operation phase of the camera module, the guiding unit operates prior to the camera module. The guiding unit is integrated by the display and the single-chip microcomputer. The single-chip microcomputer is used to store the animation video data and send it to the display in real time. The display receives the animation video data and plays the animation video data.
[0014] There is only one set of dynamic targets in the animation stored in the single chip microcomputer. When the guiding unit plays the animation, relative to the global animation image, only the dynamic targets are in motion. The moving targets are the gaze targets provided by the guiding unit to the user, and the motion trajectories of the dynamic targets are circular motions in the animation video data.
[0015] Furthermore, during the animation playing stage of the guiding unit, the user's eyes track the target, and the camera module synchronously operates to continuously collect the user's eye image data, so that the time stamp interval corresponding to the earliest group of collected user's eye image data and the last group of collected user's eye image data is equal to the time consumed by the dynamic target in the animation video data to run a circle;
[0016] Among them, the camera module is integrated with a high-definition high-speed camera. When the camera module collects user eye image data, it continuously collects user eye image data based on a specified collection frequency, so that the collected user eye image data is no less than one hundred and eighty groups.
[0017] Furthermore, when the receiving module receives the user's eye image data, it synchronously performs marking and grayscale image conversion operations on the user's eye image data;
[0018] When the receiving module performs marking processing on the user's eye image data, the user's eye image data acquisition timestamp is used for marking. When the receiving module performs grayscale image conversion on the user's eye image data, it complies with:
[0019] ;
[0020] Where: The grayscale value of the image after the user's eye image data is converted into a grayscale image; , , The color value of the user's eye image based on the RGB three channels;
[0021] After the receiving module completes the marking of the user's eye image data and the grayscale image conversion operation, the grayscale image with the mark is stored synchronously.
[0022] Furthermore, the segmentation module is provided with submodules at the lower level, including:
[0023] A selection unit, used for selecting the user's eye image data converted into a grayscale image stored in the receiving module, and feeding it back to the segmentation module;
[0024] A transmission unit, used for receiving the user eyeball image segmented by the segmentation module, and synchronously transmitting the user eyeball image to the receiving module;
[0025] Among them, when the selection unit selects the user eye image data converted into a grayscale image, the image data is selected in sequence based on the order of the image data marking timestamps, and the number of image data selected each time is 1. Each group of user eye image data converted into a grayscale image is processed based on the segmentation module to obtain two groups of user eye images. The two groups of eye images are further distinguished by the left eyeball and the right eyeball, and are further marked with the acquisition timestamp corresponding to the mark of the user eye image data from the user eye image source. When the transmission module transmits the user eye image to the receiving module for storage, it identifies the storage position of the corresponding user eye image data in the receiving module based on the acquisition timestamp marked by the user eye image, and iterates the user eye image data with the user eye image with the same acquisition timestamp.
[0026] Furthermore, during the operation phase of the segmentation module, a grayscale threshold of the user's eye image is set, and each pixel in the user's eye image data converted into a grayscale image is compared based on the grayscale threshold of the user's eye image, and pixels that do not meet the grayscale threshold of the user's eye image are taken as segmentation targets, and the user's eye image data converted into a grayscale image is segmented, and the unsegmented image area is the user's eye image;
[0027] Among them, the segmentation module runs the segmentation to obtain two groups of user eye images, and the transmission unit transmits the two groups of user eye images to the receiving module. The selection unit refreshes and runs, and the segmentation module runs again synchronously with the refresh operation of the selection unit, and the transmission unit runs again synchronously with the operation of the segmentation module.
[0028] Furthermore, the reconstruction module receives the user eyeball image from a receiving module, and the reconstruction module selects a user eyeball image receiving target based on the user eyeball image tag each time it runs, so that the user eyeball image received each time is from the left eyeball or the right eyeball;
[0029] The construction logic of the user's eye plane view in the reconstruction module is:
[0030] Traverse the user eyeball images received by all reconstruction modules, set the grayscale judgment thresholds of the cornea, iris and pupil regions, identify the centers of the cornea, iris and pupil regions in each group of user eyeball images based on the judgment thresholds, mark the acquisition timestamp sequence based on the user eyeball images, overlay the user eyeball image corresponding to the next acquisition timestamp on the user eyeball image corresponding to the previous acquisition timestamp, and make the centers of the cornea, iris and pupil regions of the mutually overlaid user eyeball images overlap;
[0031] Among them, each user's eye image is configured with a corresponding layer, and the image presented when the user's eye images on all layers are displayed synchronously is the user's eye plane view constructed by the reconstruction module.
[0032] Furthermore, the storage module stores two sets of user eyeball plane images each time, corresponding to the left eyeball and the right eyeball respectively, and the two sets of user eyeball plane images are further distinguished and marked based on the left eyeball and the right eyeball;
[0033] Among them, the storage module is internally provided with submodules, including:
[0034] The interactive unit is used to monitor whether the storage module has completed the storage operation of the user's eye plane image. After monitoring that the storage module has completed the storage operation of the user's eye plane image, the receiving module is controlled to run again. When the receiving module runs again, all image data stored internally are deleted.
[0035] Furthermore, the evaluation module is triggered to run when there are at least four groups of user eyeball plane images stored in the storage module, and the difference calculation formula of the user eyeball plane images in the evaluation module is:
[0036] ;
[0037] Where: The difference of the user's eye plane image; The maximum grayscale in the user's eyeball plane image of the latest left eyeball; is the maximum grayscale in the previous set of left eyeball user eyeball images compared to the latest left eyeball user eyeball image; The minimum grayscale in the user's eyeball plane image of the latest left eyeball; is the minimum grayscale in the previous set of left eyeball user eyeball images compared to the latest left eyeball user eyeball image;
[0038] The evaluation module runs continuously to obtain , the differences of the latest three groups of user eyeball plane images are recorded as , or , it means the user's eye disease has worsened or improved.
[0039] Furthermore, the camera module is interactively connected with a guiding unit via a wireless network, the camera module is interactively connected with a receiving module and a segmentation module via a wireless network, the segmentation module is interactively connected with a selection unit and a transmission unit via a wireless network, the selection unit and the transmission unit are interactively connected with the receiving module via a wireless network, the segmentation module is interactively connected with a reconstruction module via a wireless network, the reconstruction module is interactively connected with the receiving module via a wireless network, the reconstruction module is interactively connected with a storage module and an evaluation module via a wireless network, and the storage module is interactively connected with an interaction unit via a homogeneous wireless network.
[0040] Compared with the known public technology, the technical solution provided by the present invention has the following beneficial effects:
[0041] The present invention provides a high-precision real-time eye movement detection system. During operation, the system guides the user's eye movement through a specific animation video, thereby collecting the user's eye image during the user's eye movement, and further constructing an eye image containing more user eye information based on the superposition of the images, thereby further providing more diagnostic reference information for medical staff based on the recording and observation of the eye image, bringing more available auxiliary reference data for the diagnosis and treatment of eye diseases, and bringing a better and more accurate diagnosis and treatment experience to patients. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0043] Figure 1It is a structural diagram of a high-precision real-time eye movement detection system. DETAILED DESCRIPTION
[0044] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0045] The present invention will be further described below in conjunction with the embodiments.
[0046] Embodiment 1:
[0047] A high-precision real-time eye movement detection system of this embodiment, such as Figure 1 As shown, including:
[0048] A camera module, used to collect user eye image data;
[0049] The camera module is internally provided with sub-modules, including:
[0050] A guiding unit, used to play animation and provide a target for users to focus on;
[0051] Among them, during the operation phase of the camera module, the guiding unit operates prior to the camera module. The guiding unit is integrated by the display and the single-chip microcomputer. The single-chip microcomputer is used to store the animation video data and send it to the display in real time. The display receives the animation video data and plays the animation video data.
[0052] There is only one set of dynamic targets in the animation stored in the single chip. When the guiding unit plays the animation, relative to the global animation image, only the dynamic targets are in motion. The motion targets are the gaze targets provided by the guiding unit to the user. The motion trajectory of the dynamic targets is a circular motion in the animation video data.
[0053] A receiving module, used for receiving the user's eye image data collected by the camera module and storing the user's eye image data;
[0054] The receiving module receives the user's eye image data and synchronously performs marking and grayscale image conversion operations on the user's eye image data;
[0055] When the receiving module performs marking processing on the user's eye image data, the user's eye image data acquisition timestamp is used for marking. When the receiving module performs grayscale image conversion on the user's eye image data, it complies with:
[0056] ;
[0057] Where: The grayscale value of the image after the user's eye image data is converted into a grayscale image; , , The color value of the user's eye image based on the RGB three channels;
[0058] After the receiving module completes the marking of the user's eye image data and the grayscale image conversion operation, the grayscale image with the mark is stored synchronously;
[0059] A segmentation module is used to obtain the user's eye image data stored in the receiving module, perform segmentation processing on the user's eye image data to obtain the user's eyeball image and feed it back to the receiving module for storage;
[0060] The segmentation module is provided with submodules at the lower level, including:
[0061] A selection unit, used for selecting the user's eye image data converted into a grayscale image stored in the receiving module, and feeding it back to the segmentation module;
[0062] A transmission unit, used for receiving the user eyeball image segmented by the segmentation module, and synchronously transmitting the user eyeball image to the receiving module;
[0063] Wherein, when the selection unit selects the user eye image data converted into a grayscale image, the image data is selected in sequence based on the order of the image data marking timestamps, and the number of image data selected each time is 1. Each group of user eye image data converted into a grayscale image is processed based on the segmentation module to obtain two groups of user eye images. The two groups of eye images are further distinguished by left eyeballs and right eyeballs, and are further marked with the acquisition timestamp corresponding to the mark of the user eye image data of the user eye image source. When the transmission module transmits the user eye image to the receiving module for storage, the transmission module identifies the storage location of the corresponding user eye image data in the receiving module based on the acquisition timestamp marked by the user eye image, and iterates the user eye image data with the user eye image with the same acquisition timestamp;
[0064] A reconstruction module, used for receiving a user's eyeball image and constructing a user's eyeball plan view based on the user's eyeball image;
[0065] A storage module, used for receiving the user eyeball plane image constructed in the reconstruction module, marking the user eyeball plane image with a construction timestamp and storing the image;
[0066] The storage module runs each time to store two groups of user eyeball plane images, corresponding to the left eyeball and the right eyeball respectively, and the two groups of user eyeball plane images are further distinguished and marked based on the left eyeball and the right eyeball;
[0067] Among them, the storage module is internally provided with submodules, including:
[0068] The interaction unit is used to monitor whether the storage module has completed the storage operation of the user's eyeball plane image, and after monitoring that the storage module has completed the storage operation of the user's eyeball plane image, control the receiving module to run again, and when the receiving module runs again, delete all the image data stored internally;
[0069] An evaluation module, used for receiving the user's eyeball plane image stored in the storage module, and evaluating the difference of the user's eyeball plane image;
[0070] The camera module is interactively connected with a guiding unit via a wireless network, the camera module is interactively connected with a receiving module and a segmentation module via a wireless network, the segmentation module is interactively connected with a selection unit and a transmission unit via a wireless network, the selection unit and the transmission unit are interactively connected with the receiving module via a wireless network, the segmentation module is interactively connected with a reconstruction module via a wireless network, the reconstruction module is interactively connected with the receiving module via a wireless network, the reconstruction module is interactively connected with a storage module and an evaluation module via a wireless network, and the storage module is interactively connected with an interaction unit via a homogeneous wireless network.
[0071] In this embodiment, the camera module operates to collect user eye image data, the guiding unit synchronously plays an animation to provide the user with a gaze target, the receiving module is rear-positioned to receive the user eye image data collected by the camera module, and stores the user eye image data, the segmentation module further obtains the user eye image data stored in the receiving module, performs segmentation processing on the user eye image data to obtain the user eye image and feeds it back to the receiving module for storage again, the selection unit synchronously selects the user eye image data converted into a grayscale image stored in the receiving module, and feeds it back to the segmentation module, the transmission unit receives the user eye image segmented by the segmentation module in real time, and synchronously transmits the user eye image to the receiving module, and ... The receiving module transmits, and then the reconstruction module receives the user's eye image, constructs a user's eye plane view based on the user's eye image, and receives the user's eye plane image constructed in the reconstruction module through the storage module, and marks and stores the user's eye plane image with a construction timestamp. The interactive unit monitors in real time whether the storage module has completed the storage operation of the user's eye plane image. After monitoring that the storage module has completed the storage operation of the user's eye plane image, the receiving module is controlled to run again. During the stage when the receiving module runs again, all image data stored internally are deleted. Finally, the user's eye plane image stored in the storage module is received through the evaluation module, and the difference of the user's eye plane image is evaluated.
[0072] Through the operation of the system in the above embodiment, with machine vision as the core technology, the eye movement detection effect is brought about, which provides more available auxiliary reference data for the diagnosis and treatment of eye diseases.
[0073] Embodiment 2:
[0074] In terms of specific implementation, based on Example 1, this example refers to Figure 1 The high-precision real-time eye movement detection system in Example 1 is further described in detail:
[0075] During the animation playing stage of the guiding unit, the user's eyes track the target, and the camera module synchronously runs to continuously collect the user's eye image data, so that the time stamp interval corresponding to the earliest group of collected user's eye image data and the last group of collected user's eye image data is equal to the time consumed by the dynamic target in the animation video data to run a circle;
[0076] The camera module is integrated with a high-definition high-speed camera. When the camera module collects the user's eye image data, it continuously collects the user's eye image data based on a specified collection frequency, so that the collected user's eye image data is not less than one hundred and eighty groups;
[0077] In the running stage of the segmentation module, a grayscale threshold of the user's eyeball image is set, and each pixel in the user's eyeball image data converted into a grayscale image is compared based on the grayscale threshold of the user's eyeball image. Pixels that do not meet the grayscale threshold of the user's eyeball image are taken as segmentation targets, and the user's eyeball image data converted into a grayscale image is segmented. The image area that is not segmented is the user's eyeball image;
[0078] Among them, the segmentation module runs the segmentation to obtain two groups of user eye images, and the transmission unit transmits the two groups of user eye images to the receiving module. The selection unit refreshes and runs, and the segmentation module runs again synchronously with the refresh operation of the selection unit, and the transmission unit runs again synchronously with the operation of the segmentation module.
[0079] In this embodiment, through the above settings, further operation data support is provided for the operation of the system in Example 1, ensuring the stable operation of the system in Example 1 and realizing eye movement detection.
[0080] Embodiment 3:
[0081] In terms of specific implementation, based on Example 1, this example refers to Figure 1 The high-precision real-time eye movement detection system in Example 1 is further described in detail:
[0082] The reconstruction module receives the user's eyeball image from the receiving module. The reconstruction module selects the user's eyeball image receiving target based on the user's eyeball image tag each time it runs, so that the user's eyeball image received each time is from the left eyeball or the right eyeball;
[0083] The construction logic of the user's eye plane view in the reconstruction module is:
[0084] Traverse the user eyeball images received by all reconstruction modules, set the grayscale judgment thresholds of the cornea, iris and pupil regions, identify the centers of the cornea, iris and pupil regions in each group of user eyeball images based on the judgment thresholds, mark the acquisition timestamp sequence based on the user eyeball images, overlay the user eyeball image corresponding to the next acquisition timestamp on the user eyeball image corresponding to the previous acquisition timestamp, and make the centers of the cornea, iris and pupil regions of the mutually overlaid user eyeball images overlap;
[0085] Each user's eyeball image is configured with a corresponding layer, and the image presented when the user's eyeball images on all layers are displayed synchronously is the user's eyeball plane view constructed by the reconstruction module;
[0086] The evaluation module is triggered to run when there are at least four sets of user eyeball plane images stored in the storage module. The difference calculation formula of the user eyeball plane images in the evaluation module is:
[0087] ;
[0088] Where: The difference of the user's eye plane image; The maximum grayscale in the user's eyeball plane image of the latest left eyeball; is the maximum grayscale in the previous set of left eyeball user eyeball images compared to the latest left eyeball user eyeball image; The minimum grayscale in the user's eyeball plane image of the latest left eyeball; is the minimum grayscale in the previous set of left eyeball user eyeball images compared to the latest left eyeball user eyeball image;
[0089] The evaluation module runs continuously to obtain , the differences of the latest three groups of user eyeball plane images are recorded as , or , it means the user's eye disease has worsened or improved.
[0090] In this embodiment, the above-mentioned logical formula further defines the difference evaluation logic of the user's eye plane image in the evaluation module in Embodiment 1, so that the user's eye disease status is further determined based on the evaluation results.
[0091] In summary, during operation, the system in the above embodiment guides the user's eye movement through a specific animation video, thereby collecting the user's eye image during the user's eye movement, and further constructing an eye image containing more user eye information based on the superposition of the images, thereby further providing more diagnostic reference information for medical staff based on the recording and observation of the eye image, providing more available auxiliary reference data for the diagnosis and treatment of eye diseases, and bringing a better and more accurate diagnosis and treatment experience to patients.
[0092] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A high-precision real-time eye movement detection system, characterized in that: include: A camera module, used to collect user eye image data; A receiving module is used to receive the user's eye image data collected by the camera module and store the user's eye image data; a segmentation module, configured to obtain the user's eye image data stored in the receiving module, perform segmentation processing on the user's eye image data to obtain the user's eyeball image and feed it back to the receiving module for storage; a reconstruction module, configured to receive an eye image of a user and construct a planar view of the user's eye based on the eye image; a storage module, configured to receive the user's eyeball plane image constructed in the reconstruction module, mark the user's eyeball plane image with a construction timestamp, and store the image; The evaluation module is used to receive the user's eye plane image stored in the storage module and evaluate the difference of the user's eye plane image.
2. A high-precision real-time eye movement detection system according to claim 1, characterized in that: The camera module is internally provided with submodules, including: A guidance unit, used to play animation and provide a user with a focus target; Among them, during the operation phase of the camera module, the guiding unit takes precedence over the camera module. The guiding unit is integrated with the display and the single-chip microcomputer. The single-chip microcomputer is used to store the animation video data and send it to the display in real time. The display receives the animation video data and plays the animation video data. There is only one set of dynamic targets in the animation stored in the single-chip microcomputer. When the guidance unit plays the animation, relative to the global animation image, only the dynamic targets are in motion. The moving targets are the gaze targets provided to the user by the guidance unit, and the motion trajectories of the dynamic targets are circular motions in the animation video data.
3. A high-precision real-time eye movement detection system according to claim 2, characterized in that: During the animation playback phase of the guidance unit, the user's eyes track the target, and the camera module synchronously operates to continuously collect user eye image data, so that the time interval corresponding to the earliest set of collected user eye image data and the last set of collected user eye image data is equal to the time consumed by the dynamic target in the animation video data to complete one circle; Among them, the camera module is integrated by a high-definition and high-speed camera. When the camera module collects user eye image data, it continuously collects user eye image data based on a specified collection frequency, so that the collected user eye image data is no less than one hundred and eighty groups.
4. A high-precision real-time eye movement detection system according to claim 1, characterized in that: The receiving module receives the user's eye image data and synchronously performs marking and grayscale image conversion operations on the user's eye image data; When the receiving module performs marking processing on the user's eye image data, the user's eye image data acquisition timestamp is used for marking. When the receiving module performs grayscale image conversion on the user's eye image data, it complies with: ; Where: The grayscale value of the image after the user's eye image data is converted into a grayscale image; 、 、 The color value of the user's eye image based on the RGB three channels; After the receiving module completes the marking and grayscale image conversion operations on the user's eye image data, it simultaneously stores the grayscale image with the marking.
5. A high-precision real-time eye movement detection system according to claim 1, characterized in that: The segmentation module is provided with submodules at the lower level, including: A selection unit, configured to select the user eye image data converted into a grayscale image stored in the receiving module and feed the data back to the segmentation module; A transmission unit, configured to receive the user eye image segmented by the segmentation module, and synchronously transmit the user eye image to the receiving module; Among them, when the selection unit selects the user eye image data converted into a grayscale image, the image data is selected in sequence based on the order of the image data marking timestamps, and the number of image data selected each time is 1. Each group of user eye image data converted into a grayscale image is processed based on the segmentation module to obtain two groups of user eye images. The two groups of eye images are further distinguished by the left eyeball and the right eyeball, and are further marked with the acquisition timestamp corresponding to the mark of the user eye image data of the user eye image source. When the transmission module transmits the user eye image to the receiving module for storage, it identifies the storage location of the corresponding user eye image data in the receiving module based on the acquisition timestamp marked by the user eye image, and iterates the user eye image data with the user eye image with the same acquisition timestamp.
6. A high-precision real-time eye movement detection system according to claim 1 or 5, characterized in that: During the operation phase of the segmentation module, a grayscale threshold of the user's eye image is set, and pixels in the user's eye image data converted into a grayscale image are compared based on the grayscale threshold of the user's eye image. Pixels that do not meet the grayscale threshold of the user's eye image are used as segmentation targets, and the user's eye image data converted into a grayscale image is segmented. The unsegmented image area is the user's eye image; Among them, the segmentation module runs the segmentation to obtain two groups of user eye images, and the transmission unit transmits the two groups of user eye images to the receiving module. The selection unit refreshes and runs, and the segmentation module runs again synchronously following the refresh operation of the selection unit, and the transmission unit runs again synchronously following the operation of the segmentation module.
7. A high-precision real-time eye movement detection system according to claim 1, characterized in that: The reconstruction module receives the user's eye image from a receiving module, and the reconstruction module selects a user's eye image receiving target based on the user's eye image tag each time it runs, so that the user's eye image received each time is from the left eye or the right eye; The construction logic of the user's eye plane view in the reconstruction module is: Traverse all user eye images received by the reconstruction module, set grayscale determination thresholds for the cornea, iris, and pupil regions, identify the centers of the cornea, iris, and pupil regions in each group of user eye images based on the determination thresholds, mark the acquisition timestamp sequence based on the user eye images, and overlay the user eye image corresponding to the next acquisition timestamp on the user eye image corresponding to the previous acquisition timestamp, with the centers of the cornea, iris, and pupil regions of the overlaid user eye images overlapping; Among them, each user's eye image is configured with a corresponding layer, and the image presented when the user's eye images on all layers are displayed synchronously is the user's eye plane view constructed by the reconstruction module.
8. A high-precision real-time eye movement detection system according to claim 1, characterized in that: The storage module stores two sets of user eyeball plane images each time it runs, corresponding to the left eyeball and the right eyeball respectively, and the two sets of user eyeball plane images are further distinguished by marking the left eyeball and the right eyeball; The storage module is internally provided with submodules, including: The interactive unit is used to monitor whether the storage module has completed the storage operation of the user's eye plane image. After monitoring that the storage module has completed the storage operation of the user's eye plane image, the interactive unit controls the receiving module to run again. When the receiving module runs again, all image data stored internally are deleted.
9. A high-precision real-time eye movement detection system according to claim 1, characterized in that: The evaluation module is triggered to run when there are at least four sets of user eyeball plane images stored in the storage module. The difference calculation formula of the user eyeball plane images in the evaluation module is: ; Where: The difference in the user's eye plane image; The maximum grayscale of the user's eyeball plane image of the latest left eyeball; is the maximum grayscale of the user's eyeball plane images of the previous set of left eyeballs compared to the latest left eyeball user's eyeball plane image; The minimum grayscale of the user's eyeball plane image of the latest left eyeball; The minimum grayscale of the user's eyeball plane images of the previous set of left eyeballs compared to the latest left eyeball user's eyeball plane image; The evaluation module runs continuously to obtain , the differences of the latest three groups of user eye plane images are recorded as , or , it means the user's eye disease has worsened or improved.
10. A high-precision real-time eye movement detection system according to claim 1, characterized in that: The camera module is interactively connected to a guiding unit via a wireless network, the camera module is interactively connected to a receiving module and a segmentation module via a wireless network, the segmentation module is interactively connected to a selection unit and a transmission unit via a wireless network, the selection unit and the transmission unit are interactively connected to the receiving module via a wireless network, the segmentation module is interactively connected to a reconstruction module via a wireless network, the reconstruction module is interactively connected to the receiving module via a wireless network, the reconstruction module is interactively connected to the storage module and the evaluation module via a wireless network, and the storage module is interactively connected to an interaction unit via a wireless network.
