An automatic alignment acquisition system based on eye position detection

Abstract

The present invention relates to the field of image recognition technology, and particularly relates to an automatic alignment acquisition system based on eye position detection of the eyeball, including: a control terminal, which is the main control end of the system and is used to issue execution commands; a camera module, which is used to collect eye position image data of the user; an identification module, which is used to obtain the eye position image data of the user and identify the user eye position image in the user eye position image data; during the operation of the present invention, by means of specifying instructions to prompt the user to rotate the eyeball and collect the eye position image data of the user's eyes, the eye position variability and health status of the user's eyeball are analyzed and evaluated. At the same time, when analyzing and evaluating the eye position variability and health status of the user's eyeball, a digital analysis logic and a specified determination logic can be used to output the analysis and evaluation results, so as to provide system user data reference and make a more refined and reliable analysis and evaluation of the health of the user's eyeball eye position status.

Description

Technical Field

[0001] The present invention relates to the technical field of image recognition, and particularly relates to an automatic alignment acquisition system based on eyeball position detection. Background Art

[0002] There are various eye diseases, and strabismus and extraocular muscle paralysis are one of them. They are mostly caused by the dysfunction of extraocular muscles and the limited movement of extraocular muscles, resulting in non-parallel binocular visual axes, and their symptoms are more common in children.

[0003] Currently, the diagnosis of such diseases mostly relies on the acquisition of the patient's eye images, and the diagnosis is made by the physician's observation and evaluation. This method can make a relatively quick diagnosis for obvious strabismus and extraocular muscle paralysis, but for mild and minute strabismus and extraocular muscle paralysis problems, it is often impossible to accurately detect them and take corrective and treatment measures, resulting in the further aggravation of mild and minute strabismus and extraocular muscle paralysis problems, which is not conducive to the physical development of children. Summary of the Invention

[0004] In view of the above-mentioned drawbacks of the prior art, the present invention provides an automatic alignment acquisition system based on eyeball position detection, which solves the technical problems raised in the above background art.

[0005] To achieve the above objectives, the present invention is realized through the following technical solutions:

[0006] An automatic alignment acquisition system based on eyeball position detection, comprising:

[0007] A control terminal, which is the main control end of the system and is used to issue execution commands; a camera module, which is used to collect the eye position image data of the user; an identification module, which is used to obtain the eye position image data of the user and identify the user's eye position image in the eye position image data of the user; an analysis module, which is used to receive the user's eye position image identified by the identification module and analyze the abnormality of the user's eyeball position based on the user's eye position image; an evaluation module, which is used to obtain the abnormality of the user's eyeball position analyzed by the analysis module and evaluate the health status of the user's eyeball based on the abnormality of the user's eyeball position; a feedback module, which is used to receive the evaluation result of the user's eyeball health status in the evaluation module and feedback the evaluation result to the control terminal for storage in the control terminal.

[0008] The user on the system side accesses the system through a local area network and reads the evaluation result feedback by the feedback module in the control terminal.

[0009] Furthermore, a sub-module is provided under the camera module, including:

[0010] A prompt unit, which is used to issue a prompt audio to prompt the user to control the eyeball to rotate.

[0011] A storage unit for receiving the eye position image data collected by the camera module and storing the eye position image data;

[0012] Among them, the prompt unit stores prompt audio, and the prompt audio includes: look straight ahead, look up, look down, look left, look right, look upper left, look upper right, look lower left, look lower right. The prompt unit continuously plays the prompt audio based on a specified frequency. The user rotates the eyeballs according to the instructions of the prompt audio to make corresponding actions and hold them. After each prompt audio is broadcast, the camera module collects the user's eye position image data, and after distinguishing and marking the collected eye position image data based on the corresponding text information of the source prompt audio, the storage unit completes the storage.

[0013] Furthermore, the prompt audio issued by the prompt unit also includes: close your eyes. The prompt audio of close your eyes is broadcast every time the camera module completes the collection of the user's eye position image data. After the user completes the action of closing the eyes according to the prompt audio of close your eyes, the camera module collects the user's eye image data again and stores it in the storage unit in a bound manner with the previously collected user's eye position image data.

[0014] Furthermore, a sub-module is provided inside the recognition module, including:

[0015] An extraction unit for retrieving the user's eye position image data and extracting the user's eye position image from the user's eye position image data;

[0016] Among them, when the extraction unit retrieves the user's eye position head image data, it uses the user's eye position image data stored in the storage unit as the retrieval target. Each time the extraction unit runs, it retrieves a set of user's eye position head image data and its bound eye image data, and applies any one of the absolute difference sum algorithm or the mean square error algorithm to analyze the difference region between the two images. The difference region in the user's eye position head image data analyzed is the user's eye position image extracted by the extraction unit. The user's eye position image extracted by the extraction unit is fed back to the recognition module in real time, and the operation of refreshing and running to retrieve the user's eye position head image data again and extracting the user's eye position image from the user's eye position head image data is performed until all the user's eye position image data stored in the storage unit has been processed by the extraction unit and then ends.

[0017] Furthermore, after the recognition module recognizes the user's eye position image, it performs optimization processing on the user's eye position image. The optimization processing logic of the user's eye position image is expressed as:

[0018] ;

[0019] In the formula: 、 、 are the red, green, and blue channel values of the optimized user eye position image; , , are the original red, green, and blue channel values of the user eye position image; is the color enhancement coefficient; is the sharpness enhancement coefficient;

[0020] Among them, , , after obtaining, apply , , to iteratively update the original red, green, and blue channel values of the user eye position image, and complete the output of the optimized user eye position image.

[0021] Furthermore, a sub-module is set under the analysis module, including:

[0022] A segmentation unit for obtaining the user eye position image received by the analysis module, identifying the center of the user eye position image, and segmenting and obtaining the user pupil image from the user eye position image;

[0023] Among them, the user eye position image consists of two groups of sub-images, which respectively correspond to the left and right eyes of the user. When the segmentation unit identifies the center of the user eye position image, the two groups of sub-images are used as the recognition targets, and the centers of the two groups of user eye position images are recognized.

[0024] Furthermore, when the segmentation unit segments and obtains the user pupil image from the user eye position image, a group of pixel blocks are calibrated in the sub-image of the user eye position image as the pixel blocks representing the color feature vector of the user pupil, and then the pixel blocks with the same color feature vector are searched in the sub-image. The segmentation unit then uses the un-searched pixel blocks as the segmentation target to segment the sub-image, and the remaining area image after segmentation is recorded as the user pupil image. After the segmentation unit obtains the user pupil image, it further identifies the center of the user pupil image;

[0025] Among them, the center of the user eye position image and the center of the user pupil image from the sub-images of the same user eye position image are connected to form a group of line segments.

[0026] Furthermore, the analysis logic of the user eye position variability in the analysis module is expressed as:

[0027] k = ∑ i = 1 n [ | S a − S b | i × | L a − L b | i ] ;

[0028] In the formula: is the user eye position variability performance value; is the set of user eye position images; For one set of sub-images in the user's eye position image, it is the slope of the line segment formed by the center of the user's eye position image and the center of the user's pupil image; For the other set of sub-images in the user's eye position image, it is the slope of the line segment formed by the center of the user's eye position image and the center of the user's pupil image; For one set of sub-images in the user's eye position image, it is the length of the line segment formed by the center of the user's eye position image and the center of the user's pupil image; For the other set of sub-images in the user's eye position image, it is the length of the line segment formed by the center of the user's eye position image and the center of the user's pupil image;

[0029] Among them, is the absolute value of the difference corresponding to the i-th group of user eye position images and ; is the absolute value of the difference corresponding to the i-th group of user eye position images and . The greater the value of the user's eye position variability performance value , the worse the consistency of the synchronous transformation of the user's two eyeball eye positions. On the contrary, it indicates that the consistency of the synchronous transformation of the user's two eyeball eye positions is higher.

[0030] Furthermore, a determination threshold for the health state of the user's eyeball is set in the evaluation module. After the evaluation module obtains the analysis result of the user's eyeball eye position variability, it further compares the analysis result with the determination threshold for the health state of the user's eyeball, and evaluates whether the state of the user's eyeball is healthy based on the comparison result.

[0031] Furthermore, the control terminal is interconnected through a local area network with a camera module. The lower level of the camera module is interconnected through a local area network with a prompt unit and a storage unit. The camera module is interconnected through a local area network with an identification module. Inside the identification module, an extraction unit is electrically connected through a medium. The extraction unit is interconnected with the storage unit through a local area network. The identification module is interconnected through a local area network with an analysis module, an evaluation module, and a feedback module. The lower level of the analysis module is interconnected through a local area network with a segmentation unit.

[0032] Adopting the technical solution provided by the present invention, compared with the known public technology, it has the following beneficial effects:

[0033] The present invention provides an automatic alignment acquisition system based on eye position detection. During the operation of the system, by means of specifying instructions to prompt the user to rotate the eyes and collect the eye position image data of the user, the eye position variability and health status of the user's eyes are analyzed and evaluated. At the same time, when analyzing and evaluating the eye position variability and health status of the user's eyes, a digital analysis logic and a specified judgment logic can be used to output the analysis and evaluation results, so as to provide system user data reference, make a more refined and reliable analysis and evaluation of the health of the user's eye position state, ensure that even minor strabismus and extraocular muscle paralysis problems of the user's eye position can be detected, and thus intervene in treatment more quickly to prevent the aggravation of strabismus and extraocular muscle paralysis problems. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0035] Figure 1 It is a schematic structural diagram of an automatic alignment acquisition system based on eye position detection;

[0036] Figure 2 It is a display diagram of minor strabismus and extraocular muscle paralysis problems in the present invention;

[0037] The reference numerals in the drawings respectively represent: 1, control terminal; 2, camera module; 21, prompt unit; 22, storage unit; 3, recognition module; 31, extraction unit; 4, analysis module; 41, segmentation unit; 5, evaluation module; 6, feedback module. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0038] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0039] The following further describes the present invention with reference to the embodiments.

[0040] Embodiment 1:

[0041] An automatic alignment acquisition system based on eye position detection in this embodiment, as Figure 1 shown, includes:

[0042] The control terminal 1, which is the main control end of the system, is used to issue execution commands;

[0043] The camera module 2 is used to collect the eye position image data of the user;

[0044] The camera module 2 is provided with sub - modules at a lower level, including:

[0045] The prompt unit 21 is used to emit a prompt audio to prompt the user to control the eye rotation;

[0046] The storage unit 22 is used to receive the eye position image data collected by the camera module 2 and store the eye position image data;

[0047] Among them, the prompt unit 21 stores the prompt audio. The prompt audio includes: look straight ahead, look up, look down, look left, look right, look upper left, look upper right, look lower left, look lower right. The prompt unit 21 continuously plays the prompt audio based on a specified frequency. The user rotates the eyes according to the prompt audio instructions and makes corresponding actions and holds. After each prompt audio is broadcast, the camera module 2 collects the eye position image data of the user, and after differentiating and marking the collected eye position image data based on the corresponding text information of the source prompt audio, the storage unit 22 completes the storage;

[0048] The recognition module 3 is used to obtain the eye position image data of the user and recognize the user's eye position image in the eye position image data of the user;

[0049] The recognition module 3 is internally provided with sub - modules, including:

[0050] The extraction unit 31 is used to retrieve the eye position image data of the user and extract the user's eye position image from the eye position image data of the user;

[0051] Among them, when the extraction unit 31 retrieves the user's eye position head image data, it uses the eye position image data stored in the storage unit 22 as the retrieval target. Each time the extraction unit 31 runs, it retrieves a set of user's eye position head image data and its bound eye image data, and applies any one of the absolute difference sum algorithm or the mean square error algorithm to analyze the difference region between the two images. The difference region in the user's eye position head image data analyzed is the user's eye position image extracted by the extraction unit 31. The user's eye position image extracted by the extraction unit 31 is fed back to the recognition module 3 in real - time, and the operation of refreshing and running to retrieve the user's eye position head image data again and extracting the user's eye position image from the user's eye position head image data is performed until all the eye position image data stored in the storage unit 22 has been processed by the extraction unit 31 and then ends;

[0052] An analysis module 4, configured to receive the user eye position image recognized by the recognition module 3, and analyze the anisotropy of the user's eye position based on the user eye position image;

[0053] The analysis module 4 is provided with sub-modules at a lower level, including:

[0054] A segmentation unit 41, configured to obtain the user eye position image received by the analysis module 4, identify the center of the user eye position image, and segment and obtain the user pupil image from the user eye position image;

[0055] Wherein, the user eye position image is composed of two sets of sub-images, which respectively correspond to the user's left eye and right eye. When the segmentation unit 41 identifies the center of the user eye position image, the two sets of sub-images are used as the recognition targets, and the centers of the two sets of user eye position images are identified;

[0056] An evaluation module 5, configured to obtain the anisotropy of the user's eye position analyzed by the analysis module 4, and evaluate the health status of the user's eye based on the anisotropy of the user's eye position;

[0057] A feedback module 6, configured to receive the evaluation result of the user's eye health status in the evaluation module 5, feedback the evaluation result to the control terminal 1, and store it in the control terminal 1;

[0058] The user on the system side accesses the system through a local area network, and reads the evaluation result feedback by the feedback module 6 in the control terminal 1;

[0059] The control terminal 1 is interconnected with a camera module 2 through a local area network. The lower level of the camera module 2 is interconnected with a prompt unit 21 and a storage unit 22 through a local area network. The camera module 2 is interconnected with a recognition module 3 through a local area network. An extraction unit 31 is electrically connected inside the recognition module 3 through a medium. The extraction unit 31 is interconnected with the storage unit 22 through a local area network. The recognition module 3 is interconnected with an analysis module 4, an evaluation module 5 and a feedback module 6 through a local area network. The lower level of the analysis module 4 is interconnected with a segmentation unit 41 through a local area network.

[0060] In this embodiment, the control terminal 1 controls the camera module 2 to operate and collect the eye position image data of the user. The prompting unit 21 simultaneously emits a prompting audio to prompt the user to control the eye movement. The storage unit 22 receives in real time the eye position image data collected by the camera module 2 and stores the eye position image data. The recognition module 3 further obtains the eye position image data of the user, recognizes the user eye position image in the user eye position image data, and the extraction unit 31 simultaneously retrieves the user eye position image data and extracts the user eye position image from the user eye position image data. The analysis module 4 runs later to receive the user eye position image recognized by the recognition module 3, analyzes the anisotropy of the user's eye position based on the user eye position image, and the segmentation unit 41 simultaneously obtains the user eye position image received by the analysis module 4, recognizes the center of the user eye position image, and segments and obtains the user pupil image from the user eye position image. Then, the evaluation module 5 obtains the anisotropy of the user's eye position analyzed by the analysis module 4, evaluates the health status of the user's eye based on the anisotropy of the user's eye position, and finally the feedback module 6 receives the evaluation result of the user's eye health status in the evaluation module 5, feeds back the evaluation result to the control terminal 1, and stores it in the control terminal 1;

[0061] Through the operation of the system in the above embodiment, an intelligent monitoring method is brought to the problems of strabismus and extraocular muscle paralysis in eye diseases, so that the problems of strabismus and extraocular muscle paralysis that are small and not easily observed by humans can be discovered, so as to make timely responses and reduce the possibility of exacerbation of small strabismus and extraocular muscle paralysis problems.

[0062] Embodiment 2:

[0063] At the specific implementation level, on the basis of Embodiment 1, this embodiment refers to Figure 1 to further specifically describe an automatic alignment acquisition system based on eye position detection in Embodiment 1:

[0064] The prompting audio emitted by the prompting unit 21 also includes: please close your eyes. The prompting audio of please close your eyes is broadcast every time the camera module 2 completes the acquisition of the user's eye position image data. After the user completes the closing action according to the prompting audio of please close your eyes, the camera module 2 collects the user's eye image data again and stores it in the storage unit 22 in a bound manner with the previously collected user eye position image data.

[0065] Through the above, the operation configuration and operation logic of the prompting unit 21 are further defined.

[0066] As Figure 1 shown, after the recognition module 3 recognizes the user eye position image, it performs optimization processing on the user eye position image. The optimization processing logic of the user eye position image is expressed as:

[0067] ;

[0068] In the formula: , , are the red, green, and blue channel values of the optimized user eye position image; , , are the original red, green, and blue channel values of the user eye position image; is the color enhancement coefficient; is the sharpness enhancement coefficient;

[0069] Among them, , , After obtaining, apply , , to iterate the original red, green, and blue channel values of the user eye position image to complete the output of the optimized user eye position image.

[0070] Through the above settings, the user eye position image is optimized according to the specified logic to ensure the detailed image of the user eye position image, making the user eye position image for further application in the system operation more accurate, thereby improving the accuracy of the system operation output result.

[0071] As Figure 1 shown, when the segmentation unit 41 segments and obtains the user pupil image from the user eye position image, a set of pixel blocks are calibrated in the sub-image of the user eye position image as the pixel blocks representing the user pupil color feature vector, and then the pixel blocks with the same color feature vector are searched in the sub-image. The segmentation unit 41 then uses the pixel blocks that have not been found as the segmentation target to segment the sub-image, and the remaining area image after segmentation is recorded as the user pupil image. After the segmentation unit 41 obtains the user pupil image, it further identifies the center of the user pupil image;

[0072] Among them, the center of the user eye position image and the center of the user pupil image from the sub-image of the same user eye position image are connected to form a set of line segments.

[0073] Through the above settings, data for evaluating the health status of the user's eyeball is further obtained from the user eye position image, providing necessary operation data support for the operation of subsequent modules in the system.

[0074] Embodiment 3:

[0075] At the specific implementation level, based on Embodiment 1, this embodiment further specifically describes a kind of automatic alignment acquisition system based on eyeball eye position detection in Embodiment 1 with reference to Figure 1 :

[0076] The analysis logic of the user eyeball eye position variability in the analysis module 4 is expressed as:

[0077] k = ∑ i = 1 n [ | S a − S b | i × | L a − L b | i ] ;

[0078] In the formula: is the user's eye position variability performance value; is the set of the user's eye position images; is the slope of the line formed by the center of the user's eye position image and the center of the user's pupil image in one set of sub-images of the user's eye position images; is the slope of the line formed by the center of the user's eye position image and the center of the user's pupil image in the other set of sub-images of the user's eye position images; is the length of the line formed by the center of the user's eye position image and the center of the user's pupil image in one set of sub-images of the user's eye position images; is the length of the line formed by the center of the user's eye position image and the center of the user's pupil image in the other set of sub-images of the user's eye position images;

[0079] Among them, is the absolute value of the difference corresponding to the and in the i-th group of the user's eye position images; is the absolute value of the difference corresponding to the and in the i-th group of the user's eye position images. The larger the user's eye position variability performance value , the worse the consistency of the synchronous transformation of the user's two eye positions. On the contrary, it indicates that the consistency of the synchronous transformation of the user's two eye positions is higher;

[0080] A user's eye health status determination threshold is set in the evaluation module 5. After obtaining the analysis result of the user's eye position variability, the evaluation module 5 further compares the analysis result with the user's eye health status determination threshold, and evaluates whether the user's eye status is healthy based on the comparison result.

[0081] Through the above settings, the analysis logic of the user's eye position variability is further defined, making a determination for the operation of the evaluation module 5 and providing necessary data support.

[0082] In summary, during the operation of the system in the above embodiments, by means of specifying instructions to prompt the user to rotate the eyes and collecting the eye position image data of the user's eyes, the eye position variability and health status of the user's eyes are analyzed and evaluated. At the same time, when analyzing and evaluating the eye position variability and health status of the user's eyes, a digital analysis logic and a specified judgment logic can be used to output the analysis and evaluation results, so as to provide a reference for the system user data, make a more refined and reliable analysis and evaluation of the health of the user's eye position state, ensure that minor strabismus and extraocular muscle paralysis problems in the user's eye position can be detected, and thus intervene in the treatment more quickly to prevent the aggravation of strabismus and extraocular muscle paralysis problems.

[0083] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these 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 various embodiments of the present invention.

Claims

1. An automatic alignment acquisition system based on eye position detection, characterized in that: include: The control terminal (1) is the main control terminal of the system and is used to issue execution commands; A camera module (2), used for collecting eye position image data of a user; The recognition module (3) is used to obtain the user's eye position image data and identify the user's eye position image in the user's eye position image data; after identifying the user's eye position image, the recognition module (3) performs optimization processing on the user's eye position image. The optimization processing logic of the user's eye position image is expressed as follows: Where: R new , G new , B new are the red, green and blue channel values ​​of the optimized user eye image; R, G and B are the original red, green and blue channel values ​​of the user eye image; k1 is the color enhancement coefficient; k2 is the clarity enhancement coefficient; Among them, R ne w, G new , B new After obtaining, apply R new , G new , B new Iterate the original red, green, and blue channel values ​​of the user's eye position image to complete the output of the optimized user's eye position image; The analysis module (4) is used to receive the user eye position image identified in the identification module (3), and analyze the user's eye position variability based on the user's eye position image; the analysis logic of the user's eye position variability in the analysis module (4) is expressed as: Where: k is the variability of the user's eye position; n is the set of user eye position images; S a is the slope of the line segment formed by the center of the user eye position image of one group of sub-images and the center of the user pupil image in the two groups of sub-images of the user eye position image; S b L is the slope of the line segment formed by the center of the user eye position image of the other group of sub-images and the center of the user pupil image in the two groups of sub-images of the user eye position image; a L is the length of the line segment formed by the center of the user eye position image of one group of sub-images and the center of the user pupil image in the two groups of sub-images of the user eye position image; b is the length of the line segment formed by the center of the user eye position image of the other group of sub-images and the center of the user pupil image in the two groups of sub-images of the user eye position image; Among them, |S a -S b | i The eye position image of the i-th group of users corresponds to S a With S b The absolute value of the difference; |L a -L b | i L is the eye position image corresponding to the i-th group of users a With L b The absolute value of the difference between the two eyeballs, the larger the value k of the user's eye position variability, the worse the consistency of the synchronous transformation of the user's two eyeballs, and vice versa, the higher the consistency of the synchronous transformation of the user's two eyeballs; An evaluation module (5) is used to obtain the eye position variability of the user's eyeball analyzed by the analysis module (4), and evaluate the user's eye health status based on the eye position variability of the user's eyeball; A feedback module (6) is used to receive the evaluation result of the user's eye health status in the evaluation module (5), feed back the evaluation result to the control terminal (1), and store it in the control terminal (1); The system end user accesses the system through the local area network and reads the evaluation result fed back by the feedback module (6) in the control terminal (1).

2. The automatic alignment acquisition system based on eye position detection according to claim 1 is characterized in that: The camera module (2) is provided with a submodule at a lower level, including: A prompting unit (21), used for issuing a prompting audio to prompt the user to control eye movement; The storage unit (22) is used to receive eye position image data collected by the camera module (2) and store the eye position image data; wherein the prompt unit (21) stores prompt audio, and the prompt audio includes: please look straight ahead, please look upward, please look downward, please look to the left, please look to the right, please look to the upper left, please look to the upper right, please look to the lower left, please look to the lower right, the prompt unit (21) continuously plays the prompt audio based on a specified frequency, and the user rotates the eyeball to make corresponding movements and maintains them according to the prompt audio instructions. After each prompt audio broadcast ends, the camera module (2) collects the user's eye position image data, and distinguishes and marks the collected eye position image data based on the text information corresponding to the source prompt audio, and then the storage unit (22) completes the storage.

3. The automatic alignment acquisition system based on eye position detection according to claim 2 is characterized in that: The prompt audio emitted by the prompt unit (21) also includes: Please close your eyes. The prompt audio of "Please close your eyes" is broadcasted each time the camera module (2) completes the collection of user eye position image data. After the user completes the eye closing action according to the prompt audio of "Please close your eyes", the camera module (2) collects the user's eye image data again, and stores it in the storage unit (22) in a bound manner with the last set of user eye position image data collected.

4. The automatic alignment acquisition system based on eye position detection according to claim 1 is characterized in that: The identification module (3) is internally provided with submodules, including: An extraction unit (31) is used to retrieve user eye position image data and extract the user eye position image from the user eye position image data; When the extraction unit (31) retrieves the user eye position portrait data, the user eye position image data stored in the storage unit (22) is used as the retrieval target. Each time the extraction unit (31) runs, it retrieves a set of user eye position portrait data and its bound eye image data, and applies any one of the absolute difference algorithm and the mean square error algorithm to perform difference area analysis on the two sets of images. The difference area in the analyzed user eye position portrait data is the user eye position image extracted by the extraction unit (31). The user eye position image extracted by the extraction unit (31) is fed back to the recognition module (3) in real time, and refreshes and runs again to retrieve the user eye position portrait data, and extract the user eye position image from the user eye position portrait data, until all the user eye position image data stored in the storage unit (22) are processed by the extraction unit (31) and the operation is terminated.

5. The automatic alignment acquisition system based on eye position detection according to claim 1 is characterized in that: The analysis module (4) is provided with submodules at the lower level, including: A segmentation unit (41) is used to obtain the user eye position image received by the analysis module (4), identify the center of the user eye position image, and segment the user eye position image to obtain the user pupil image; The user eye position image is composed of two groups of sub-images, and the two groups of sub-images correspond to the left eye and the right eye of the user respectively. When the segmentation unit (41) identifies the center of the user eye position image, the two groups of sub-images are used as identification targets to identify the centers of the two groups of user eye position images.

6. The automatic alignment acquisition system based on eye position detection according to claim 5, characterized in that: When the segmentation unit (41) segments the user's eye position image to obtain the user's pupil image, it marks a group of pixel blocks in the sub-image of the user's eye position image as pixel blocks representing the user's pupil color feature vector, and further searches for pixel blocks with the same color feature vector in the sub-image. The segmentation unit (41) then segments the sub-image using the pixel blocks that have not been found as segmentation targets, and the segmentation remaining area image is recorded as the user's pupil image. After obtaining the user's pupil image, the segmentation unit (41) further identifies the center of the user's pupil image; The center of the user's eye position image and the center of the user's pupil image, which are derived from the sub-image in the same user's eye position image, are connected to each other to form a group of line segments.

7. The automatic alignment acquisition system based on eye position detection according to claim 1 is characterized in that: The evaluation module (5) is provided with a threshold value for determining the health status of the user's eyeballs. After obtaining the analysis result of the eye position variability of the user's eyeballs, the evaluation module (5) further compares the analysis result with the threshold value for determining the health status of the user's eyeballs, and evaluates whether the user's eyeball status is healthy based on the comparison result.

8. The automatic alignment acquisition system based on eye position detection according to claim 1 is characterized in that: The control terminal (1) is interactively connected to a camera module (2) via a local area network; the camera module (2) is interactively connected to a prompt unit (21) and a storage unit (22) via a local area network; the camera module (2) is interactively connected to an identification module (3) via a local area network; the identification module (3) is electrically connected to an extraction unit (31) via a medium; the extraction unit (31) is interactively connected to the storage unit (22) via a local area network; the identification module (3) is interactively connected to an analysis module (4), an evaluation module (5) and a feedback module (6) via a local area network; the analysis module (4) is interactively connected to a segmentation unit (41) via a local area network.

Images