Visual field detection method and device based on eye movement tracking

Abstract

The invention belongs to the field of visual detection, and particularly relates to a visual field detection method based on eye movement tracking, which comprises the following steps of: displaying a static sighting mark as a sighting mark which is always watched by a detected person; obtaining eyeball micro-actions; obtaining corresponding change data according to the micro-action of the eyeball; inputting the change data and the position of the mobile sighting mark on the display screen when the eyeballs have the micro-action into a neural network to obtain whether the position of the corresponding mobile sighting mark is on the view boundary or not; when the neural network outputs and obtains that the position of the movable sighting mark corresponding to the micro-action is not on the view boundary, the movable sighting mark continues to move until the position of the movable sighting mark is on the view boundary when the eyeballs have the micro-action; acquiring the position of the moving sighting mark on the display screen when the eyeballs have micro-actions; when the neural network outputs and obtains that the position of the moving sighting mark corresponding to the micro-action is located at the view boundary, the initial position and / or the moving direction of the moving sighting mark are / is changed, and the micro-action of the eyeballs continues to be obtained to determine other view boundary points of the detected person; a visual field is formed.

Description

Technical Field

[0001] The present invention belongs to the field of visual detection, and in particular relates to a visual field detection method and device based on eye tracking. Background Art

[0002] The visual field of the human eye reflects the function of the entire retina and visual pathway, and has a great influence on visual perception. Therefore, it is of great significance to accurately detect the visual field of the human eye. Traditional visual field methods rely on subjective feedback from the test subject, so for patients who do not understand the principle of visual field testing well, the test results will have large measurement errors. Summary of the invention

[0003] The present invention is proposed based on the above-mentioned needs of the prior art. The technical problem to be solved by the present invention is to provide a visual field detection method and device based on eye tracking to accurately detect the subjective factors of visual field reduction.

[0004] In order to solve the above problems, the technical solution provided by the present invention includes:

[0005] Provided is a visual field detection method based on eye tracking, comprising: displaying visual marks, wherein the visual marks include static visual marks and moving visual marks; the static visual marks are the visual marks that the person being detected always looks at, and the moving visual marks move from an initial position along a preset direction; obtaining eyeball micro-movements, wherein the micro-movements include eyeball rotation and changes in pupil size; obtaining corresponding change data according to the eyeball micro-movements, wherein the change data include an eyeball rotation change rate and a pupil size change rate, wherein the eyeball rotation change rate is determined based on a change distance of the eyeball in two frames of images and a corresponding time period, and the pupil size change rate is determined based on a change amount of a pupil radius in two frames of images and a corresponding time period; and inputting the change data and the position of the moving visual mark on the display screen when the eyeball micro-movement occurs into a neural network. , obtaining whether the position of the moving sight mark corresponding to the micro-motion is on the boundary of the visual field, the neural network is formed based on training of a training set, the training set including corresponding change data when the position of the moving sight mark overlaps with the boundary of the visual field; when the neural network outputs that the position of the moving sight mark corresponding to the micro-motion is not on the boundary of the visual field, the moving sight mark continues to move until the position of the moving sight mark is on the boundary of the visual field when the eyeball has a micro-motion during the movement; obtaining the position of the moving sight mark on the display screen when the eyeball has a micro-motion; when the neural network outputs that the position of the moving sight mark corresponding to the micro-motion is located at the boundary of the visual field, changing the initial position and / or moving direction of the moving sight mark, and continuing to obtain the micro-motion of the eyeball to determine other boundary points of the visual field of the subject; forming a visual field based on the boundary points of the visual field.

[0006] When a certain figure suddenly appears or disappears in a fixed visual field, the eyes will produce certain corresponding stress movements, including the rotation of the eyeball and / or the zooming of the pupil, and the real-time nature of the reaction is very high. According to the above principle, the visual mark is judged whether it is located at the visual field boundary according to the micro-movement of the eyeball. However, due to the instability of the eyeball, in order to prevent the tiny movement from being considered as the visual field boundary, the present invention combines a neural network to judge the visual field boundary through the constructed system, and performs different subsequent steps according to different judgment results. Due to the real-time nature of the micro-movement of the eyeball, the scope of the visual field is more accurate.

[0007] Preferably, the stationary sight mark is set at the center of the viewing window; the initial position of the moving sight mark includes a position close to the center of the viewing window or a position close to the edge of the viewing window.

[0008] Preferably, when the initial position of the moving sight mark is close to the center of the viewing window, the preset direction is the direction away from the center of the viewing window; when the initial position of the moving sight mark is close to the edge of the viewing window, the preset direction is the direction towards the center of the viewing window.

[0009] Preferably, the field of view detection method further comprises obtaining subjective field of view information of the viewer, roughly judging the range of the field of view based on the information, and accurately deriving the range of the field of view in combination with the output result of the neural network.

[0010] The above setting can reduce the number of times the neural network is input, and the approximate position of the visual mark can be obtained based on the subjective visual field information, thereby narrowing the range of eye micro-movements, thereby improving efficiency and reducing complexity.

[0011] Preferably, the moving sight mark moves in the radial direction, and when the moving direction of the moving sight mark is changed, the angles formed between two adjacent directions are the same.

[0012] A visual field detection device based on eye tracking is also provided, which performs any visual field detection method based on eye tracking, including: a display, which displays a visual mark; a camera, which obtains eye micro-movements; and a controller, which is electrically connected to the display and the camera. When the camera obtains the micro-movement, the controller obtains the position of the moving visual mark on the display screen at this time.

[0013] Preferably, the person being tested triggers a response module, which is electrically connected to the controller. When a moving sight mark appears in the field of vision of the person being tested or the moving sight mark disappears, the rough position of the field of vision is determined by triggering the response module.

[0014] Preferably, the visual field detection device includes an observation window, and the image acquisition module includes a plurality of cameras, which are arranged on the observation window and facing the viewing position of the person being detected; the distribution of the plurality of cameras is adapted to the shape of the eyes.

[0015] Preferably, the visual field detection device comprises a lens module, the lens module is arranged in front of the display, and the observation window is arranged in front of the lens module.

[0016] Preferably, when the person being tested is at the testing position, the distance between the eye and the lens module is 8 mm to 20 mm, the upper field of view angle corresponding to the upper edge of the viewing window is greater than 55°, the lower field of view angle corresponding to the lower edge is greater than 70°, the nasal field of view angle is greater than 60°, and the temporal field of view angle should be greater than 90°.

[0017] Compared with the prior art, the present invention has found the corresponding relationship between eye movement micro-movement and visual field. When a certain figure suddenly appears or disappears in a fixed visual field area, the eyes will produce certain stress movements correspondingly, including the rotation of the eyeball and / or the zooming of the pupil, and the real-time nature of the reaction is very high. According to the above principle, the visual mark is judged whether it is located at the boundary of the visual field according to the micro-movement of the eyeball. However, due to the instability of the eyeball, in order to prevent the tiny movements from being considered as the boundary of the visual field, the present invention combines a neural network, and the boundary of the visual field is judged by the constructed system, and different subsequent steps are performed according to different judgment results. Due to the real-time nature of the eyeball micro-movement, the scope of the visual field is more accurate. In addition, by using the result of the original subjective response, the influence of unnecessary micro-movements before the subjective response result on the test result is effectively avoided, thereby improving the accuracy of the overall detection result. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of this specification. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0019] Figure 1 A flowchart of a method for visual field detection based on eye tracking according to an embodiment of the present invention;

[0020] Figure 2 This is a schematic diagram of an image of two adjacent frames representing eyeball rotation in an embodiment of the present invention;

[0021] Figure 3 Schematic diagram of eyeball rotation changes in an embodiment of the present invention;

[0022] Figure 4 Schematic diagram of the change of the oblique rotation of the eyeball in the embodiment of the present invention;

[0023] Figure 5 Schematic diagram of changes in vertical rotation of the eyeball in an embodiment of the present invention;

[0024] Figure 6 is a schematic diagram of an image of two adjacent frames representing pupil response in an embodiment of the present invention;

[0025] Figure 7 Schematic diagram of changes in pupil zoom in an embodiment of the present invention;

[0026] Figure 8 Schematic diagram of the side structure of the visual field detection device based on eye tracking in an embodiment of the present invention;

[0027] Fig. 9 Schematic diagram of the distribution of cameras on a light-transmitting support plate in an embodiment of the present invention;

[0028] Fig.10 Schematic diagram of static sight marks and moving sight marks distributed on a display in an embodiment of the present invention;

[0029] Fig.11 A schematic diagram of a moving method of moving a visual mark on a display in an embodiment of the present invention;

[0030] Fig.12 A schematic diagram of another moving method of moving a visual mark on a display in an embodiment of the present invention;

[0031] Fig.13 This is a schematic diagram of determining the visual field by moving the moving sight mark in multiple directions in an embodiment of the present invention;

[0032] Fig.14 This is a schematic diagram of another method of determining visual field by moving the sight mark in multiple directions in an embodiment of the present invention. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0034] In the description of the embodiments of the present invention, it should be noted that, unless otherwise clearly specified and limited, the term "connected" should be understood in a broad sense, for example, it can be a fixed connection, or a detachable connection, or an integral connection, which can be a mechanical connection, or an electrical connection, which can be a direct connection, or can be indirectly connected through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0035] The terms "top", "bottom", "above", "lower", and "on" used throughout the description are relative to the relative positions of components of a device, such as the relative positions of the top and bottom substrates within a device. It is understood that devices are multifunctional regardless of their orientation in space.

[0036] To facilitate understanding of the embodiments of the present application, further explanation will be given below with reference to specific embodiments in conjunction with the accompanying drawings. The embodiments do not constitute a limitation on the embodiments of the present application.

[0037] Example 1

[0038] This embodiment provides a visual field detection method based on eye tracking, such as Figure 1 shown.

[0039] Display sight marks, which include a static sight mark 4 and a moving sight mark 5; the static sight mark 4 is the sight mark that the person being tested always looks at, and the moving sight mark 5 moves from an initial position along a preset direction.

[0040] Furthermore, the person to be tested faces the viewing window, the static sight mark 4 is set at the center of the viewing window, and the person to be tested always looks at the static sight mark 4. The moving sight mark 5 moves in the viewing window, and when the initial position of the moving sight mark 5 is located near the static sight mark 4, its preset direction is the direction toward the edge of the viewing window; when the initial position of the moving sight mark 5 is located near the edge of the viewing window, its corresponding preset direction is the direction toward the center of the viewing window. Further, the preset direction is the radial direction of the viewing window.

[0041] Eye micro-movements are acquired, including eye rotations and changes in pupil size.

[0042] When the viewer is looking at the stationary sight mark 4 at the center of the viewing window, the visual field coverage is fixed. When the moving sight mark 5 takes the edge of the viewing window as the initial position, the moving sight mark 5 suddenly appears in the area covered by the visual field, which will stimulate the eyes to a certain extent. Under the external stimulation, the eyes will make a certain stress response, and the corresponding eyeball will slightly deflect to a certain extent, or the size of the pupil will slightly expand to a certain extent. The above stress reactions may also occur at the same time. Correspondingly, when the moving sight mark 5 takes the center of the viewing window as the initial position, the moving sight mark 5 in the area covered by the visual field will move at a stable speed and in a preset direction. When the moving sight mark 5 disappears in the area, the eyes will also produce a certain response, such as deflection of the eyeball and expansion of the pupil.

[0043] Corresponding change data are obtained according to the micro-movements of the eyeballs, and the change data include the eyeball rotation change rate and the pupil size change rate. The eyeball rotation change rate is determined based on the change distance of the eyeballs in the two frames of images and the corresponding time period, and the pupil size change rate is determined based on the change amount of the pupil radius in the two frames of images and the corresponding time period.

[0044] The acquisition of eyeball micro-movements is achieved through the camera 301. The camera 301 can detect the position information of the human eye in real time, and the position information includes the pupil center coordinates and the pupil radius. The change in eyeball rotation is determined by the offset of the pupil center coordinates, and the change in pupil zoom is determined by the change in the pupil radius.

[0045] The images captured by the camera 301 are arranged in frame order. Taking two adjacent frames of images as an example, the two frames are labeled Z1 and Z2 respectively. Figure 2 As shown in the figure, a possible action is shown, where for Z1, the detection target has not yet appeared in the human eye's field of vision. For Z2, the detection target appears in the human eye, and the human eye has a micro-movement, which is manifested as a horizontal rotation of the eyeball. Figure 3 As shown in the figure, by overlapping the two frames of images, the distance L of eyeball rotation can be obtained. Assuming that the shooting interval between the two frames of images is t, the rate of change of eye rotation distance over time P1 is The direction of appearance or disappearance of the moving sight mark 5 will affect the direction of eyeball rotation to a certain extent. The rotation of the eyeball may also appear as follows: Figure 4 The oblique rotation shown or Figure 5 Vertical rotation shown.

[0046] In addition, the appearance or loss of sight marks in the visual field is accompanied by the movement of shrinking through-holes, e.g. Figure 6 As shown in the figure, two frames of images are taken continuously, Z3 is the first frame, and Z4 is the second frame. When taking Z3, the sight mark has not yet appeared in the human eye, and the pupil size is normal. When taking Z4, the sight mark appears in the human eye, and the pupil shrinks. Figure 7 As shown, by merging the two frames of images, we can see that the change in pupil radius before and after the sight mark appears in the human eye is R. Assuming that the time interval between the two frames of images is t, the rate of change of the pupil radius of the human eye over time P2 is:

[0047] The change data is input into a neural network to obtain whether the position of the moving sight mark 5 corresponding to the micro-movement is on the boundary of the visual field. The neural network is formed based on a training set, and the training set includes the corresponding change data when the position of the moving sight mark 5 overlaps with the boundary of the visual field.

[0048] The training set includes at least 10,000 representative subjective visual field test data as input data sets. The acquisition process includes: installing an eye tracking component on a subjective visual field detection device, asking the subject to subjectively feedback when the visual mark is observed, recording the visual field range of the subject, and intercepting the human eye image taken by the eye tracking component before and after the subject observes the visual mark. Each piece of data needs to be preprocessed to ensure that each piece of data has no missing values ​​or abnormal values, and the data is standardized or normalized to have a unified format and numerical range. Each piece of data is set with a label to declare the visual field range corresponding to the data.

[0049] The neural network is a convolutional neural network, which contains no less than 10 layers, the number of neurons is adjusted according to the fitting results, and the activation function is a rectified linear unit function (ReLU).

[0050] Before training begins, random initial values ​​are assigned to the connection weights and neuron biases in the network. The training set is input into the constructed neural network, and the output results are obtained through calculations at each layer. The neurons in each layer receive the input of the previous layer and obtain the output of the layer through the activation function. The output of the convolutional neural network is compared with the true label, and the mean square error is used as the loss function to predict the difference between the result and the true result. According to the value of the loss function, the gradient of the parameters of each network layer to the loss is calculated by the back propagation algorithm. The stochastic gradient descent optimization algorithm is used to update the parameters of each network layer according to the calculated gradient to reduce the value of the loss function.

[0051] When the neural network outputs that the position of the moving sight mark 5 corresponding to the micro-movement is not on the boundary of the visual field, the micro-movement is considered to be an interference movement in the viewing process. At this time, the moving sight mark 5 needs to continue moving until the position of the moving sight mark 5 is on the boundary of the visual field when the eyeball shows a micro-movement during the movement.

[0052] The position of the moving sight mark 5 on the display screen is obtained when the eyeball has a micro-movement.

[0053] At this time, the visual field boundary of the person to be tested in the moving direction of the moving sight mark 5 can be obtained.

[0054] When the neural network outputs that the position of the moving sight mark 5 corresponding to the micro-movement is at the boundary of the visual field, the initial position and / or moving direction of the moving sight mark 5 is changed, and the micro-movement of the eyeball continues to be acquired to determine other boundary points of the visual field of the subject.

[0055] The entire field of view is divided into multiple different areas along the straight line passing through the center. Each area corresponds to an angle. The angles corresponding to the entire field of view are θ1, θ2, …, θ nFor example, based on the first preset direction, the corresponding visual angle in the next direction is obtained by rotating θ1. In each direction, the moving sight mark 5 has two optional directions of movement, namely Fig.14 From the edge of the screen to the center of the screen as shown, or as Fig.13 As shown, the image moves from the center of the screen to the edge of the screen. By analogy, the entire visual field is obtained. Furthermore, the entire visual field is evenly divided into regions, and the angles corresponding to each region are equal, which are all θ.

[0056] The field of view is formed based on the field of view boundary points.

[0057] Through the above process, a field of view range consisting of a plurality of field of view boundary points can be formed, which is the field of view range of the person to be detected.

[0058] Furthermore, the subjective field of view information of the viewer is obtained, and the range of the field of view is roughly determined based on the information. The micro-movements of the eyeballs that occur before the subjective information of the viewer is obtained are conveniently matched, and the output results of the neural network are combined to accurately obtain the range of the field of view.

[0059] The above setting is used to determine the boundary position of the visual field according to the real-time response of the person to be tested. Compared with the subjective response, it has good real-time performance. Since the physiological response is much higher than the subjective response speed, the demarcation range of the visual field can be accurately determined.

[0060] Example 2

[0061] This embodiment provides a visual field detection device based on eye tracking, which is used to implement the method involved in embodiment 1, such as Figure 8 shown.

[0062] The visual field detection device includes a display 1, a lens assembly, a light-transmitting support plate 302, a camera 301 and a controller.

[0063] The display 1 is used to display visual marks, and the display 1 is a flat display 1, which displays images and emits light for viewers to watch. Fig.10 As shown, the display 1 displays visual marks, which include a static visual mark 4 and a moving visual mark 5. The static visual mark 4 is displayed at the center of the display 1, i.e., corresponding to the center of the visual field of the person being tested, and the moving visual mark 5 moves continuously along a certain trajectory during the testing process.

[0064] Furthermore, the viewer's perception of brightness can be detected by adjusting the brightness of the sight mark, and different sight mark brightnesses can be adjusted according to different viewers to better personalize and adapt to the actual needs of different viewers.

[0065] The lens module 2 may be a single lens or a lens group formed by a plurality of lenses. The lens module 2 is arranged opposite to the display 1, and the light emitted by the display 1 passes through the lens module 2 and enters the eyes of the person to be detected.

[0066] The light-transmitting support plate 302 is arranged as a viewing window opposite to the lens module 2. A camera 301 is arranged on the light-transmitting support plate 302, and the camera 301 is arranged on the side facing the viewer.

[0067] The camera 301 is set toward the eyes of the person to be tested, and the micro-movements of the eyes of the person to be tested are obtained in real time. Further, in order to ensure the accuracy of the test results, the camera 301 includes multiple, exemplary, Fig. 9 As shown, there are eight cameras 301 , which are evenly arranged around the center in a shape similar to an eye.

[0068] Furthermore, the visual field detection device also includes a response module, and the subject presses the response module to record the visual mark he sees. The data will be analyzed with the micro-movements obtained by the micro camera 301 as influencing factors to form eye movement data, and the visual field range of the subject can be calculated.

[0069] The controller is electrically connected to the display 1, the camera 301 and the response module respectively. The controller controls the display 1 to display the sight mark and controls the position and moving direction of the moving sight mark 5. When the controller receives the micro-movement obtained by the camera 301, the controller obtains the position of the moving sight mark 5 on the display screen at this time. When the controller receives the response of the response module, it can judge the approximate visual field range of the person being detected according to the response. That is, the controller can implement the visual field detection method in Example 1.

[0070] The visual field detection device has a viewing window and a viewing position, the distance between the viewing position and the lens module 2 ranges from 8mm to 20mm, the upper field of view angle corresponding to the upper edge of the viewing window is greater than 55°, the lower field of view angle corresponding to the lower edge is greater than 70°, the nasal field of view angle is greater than 60°, and the temporal field of view angle should be greater than 90°.

[0071] When performing visual field detection, the display 1 displays the sight mark, the viewer faces the viewing window at the viewing position, and the image acquisition module starts working. The viewer is required to always look at the static sight mark 4 located in the center of the viewing window and try to keep it as much as possible during the visual field detection process. The moving sight mark 5 moves in a certain direction, and the viewer receives the moving sight mark 5 through the peripheral vision. When the viewer receives the appearance or disappearance of the moving sight mark 5, the eyes produce stress micro-movements, and the image acquisition module obtains the corresponding stress micro-movements, and at the same time, the corresponding visual field data at the position of the moving sight mark 5 at this time is obtained.

[0072] Further, the moving sight mark 5 moves along the preset straight line, including: Fig.11 The moving sight mark 5 shown in FIG. 1 moves from the edge of the screen to the center of the screen along a preset straight line, or as shown in FIG. Fig.12 The moving sight mark 5 shown moves from the center of the screen to the edge of the screen along a preset straight line. Specifically, the initial position of the moving sight mark 5 can be located in the area of ​​the edge of the screen, and moves along the preset straight line to the center of the screen. In this case, the viewer cannot see the moving sight mark 5 at the beginning, and when the moving sight mark 5 moves to a certain position, it can be received by the viewer. The initial position of the moving sight mark 5 can be located in the area of ​​the center of the screen, and moves along the preset straight line to the edge of the screen. In this case, the viewer can see the moving sight mark 5 at the beginning, and when the moving sight mark 5 moves to a certain position, the viewer cannot receive the moving sight mark 5.

[0073] During the movement, when the moving sight mark 5 appears in the receiving range of the viewer's eyes or disappears in the receiving range of the viewer's eyes, the eyes will be accompanied by an excited action, such as a change in pupil size, a slight turn of the eyeball, etc. When the image acquisition module captures the corresponding eye movement, it can record and determine the position of the corresponding moving sight mark 5 when the eyeball has a slight movement, thereby obtaining the corresponding visual angle. Exemplarily, when the moving sight mark 5 moves in a straight line along a first preset direction, the corresponding visual angle in the first preset direction can be obtained.

[0074] The entire field of view is divided into multiple different areas along the straight line passing through the center. Each area corresponds to an angle. The angles corresponding to the entire field of view are θ1, θ2, …, θ n For example, based on the first preset direction, the corresponding visual angle in the next direction is obtained by rotating θ1. In each direction, the moving sight mark 5 has two optional directions of movement, namely Fig.14 From the edge of the screen to the center of the screen as shown, or as Fig.13 As shown, the image moves from the center of the screen to the edge of the screen. By analogy, the entire visual field is obtained. Furthermore, the entire visual field is evenly divided into regions, and the angles corresponding to each region are equal.

[0075] The specific implementation methods described above further illustrate the purpose, technical solutions and beneficial effects of the present application in detail. It should be understood that the above description is only the specific implementation method of the present application and is not intended to limit the scope of protection of the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A visual field detection method based on eye tracking, characterized in that: include: Displaying sight marks, wherein the sight marks include static sight marks and moving sight marks; The stationary sight mark is the sight mark that the person being tested always looks at, and the moving sight mark moves from an initial position along a preset direction; Acquiring eye micro-movements, wherein the micro-movements include eye rotations and changes in pupil size; Obtaining corresponding change data according to the micro-movement of the eyeball, the change data including the eyeball rotation change rate and the pupil size change rate, the eyeball rotation change rate is determined based on the change distance of the eyeball in the two frames of images and the corresponding time period, and the pupil size change rate is determined based on the change amount of the pupil radius in the two frames of images and the corresponding time period; Inputting the change data and the position of the moving sight mark on the display screen when the eyeball has a micro-movement into a neural network to obtain whether the position of the moving sight mark corresponding to the micro-movement is on the boundary of the visual field, wherein the neural network is formed based on a training set, and the training set includes the change data corresponding to when the position of the moving sight mark overlaps with the boundary of the visual field; When the neural network outputs that the position of the moving sight mark corresponding to the micro-movement is not on the boundary of the visual field, the moving sight mark continues to move until the position of the moving sight mark is on the boundary of the visual field when the eyeball has a micro-movement during the movement; Obtaining the position of the moving sight mark on the display screen when the eyeball has micro-movement; When the neural network outputs that the position of the moving sight mark corresponding to the micro-movement is located at the boundary of the visual field, the initial position and / or moving direction of the moving sight mark is changed, and the micro-movement of the eyeball is continuously acquired to determine other boundary points of the visual field of the subject; The field of view is formed based on the field of view boundary points.

2. The visual field detection method based on eye tracking according to claim 1, characterized in that: The static sight mark is set at the center of the viewing window; The initial position of the moving sight mark includes a position close to the center of the viewing window or a position close to the edge of the viewing window.

3. The visual field detection method based on eye tracking according to claim 1, characterized in that: When the initial position of the moving sight mark is close to the center of the viewing window, the preset direction is a direction away from the center of the viewing window; When the initial position of the moving sight mark is close to the edge of the viewing window, the preset direction is the direction toward the center of the viewing window.

4. The visual field detection method based on eye tracking according to claim 1, characterized in that: The field of view detection method also includes obtaining subjective field of view information of the viewer, roughly determining the range of the field of view based on the information, and accurately deriving the field of view range in combination with the output result of the neural network.

5. The visual field detection method based on eye tracking according to claim 1, characterized in that: The moving sight mark moves radially, and when the moving direction of the moving sight mark is changed, the angles formed between two adjacent directions are the same.

6. A visual field detection device based on eye tracking, performing a visual field detection method based on eye tracking as claimed in any one of claims 1 to 5, characterized in that: include: A display, displaying sight marks; Camera, to capture eye micro-movements; The controller is electrically connected to the display and the camera. When the camera obtains a micro-motion, the controller obtains the position of the moving sight mark on the display screen at this time.

7. The visual field detection device based on eye tracking according to claim 6, characterized in that: The detected person triggers a response module, which is electrically connected to the controller. When a moving sight mark appears in the detected person's field of vision or the moving sight mark disappears, the rough position of the field of vision is determined by triggering the response module.

8. The visual field detection device based on eye tracking according to claim 6, characterized in that: The visual field detection device comprises an observation window, and the visual field detection device comprises a plurality of cameras, which are arranged on the observation window and facing the viewing position of the detected person; the distribution of the plurality of cameras is adapted to the shape of the eyes.

9. The visual field detection device based on eye tracking according to claim 6, characterized in that: The visual field detection device comprises a lens module, the lens module is arranged in front of the display, and an observation window is arranged in front of the lens module.

10. The visual field detection device based on eye tracking according to claim 6, characterized in that: When the person being tested is at the testing position, the distance between the eye and the lens module is 8mm to 20mm, the upper field of view angle corresponding to the upper edge of the viewing window is greater than 55°, the lower field of view angle corresponding to the lower edge is greater than 70°, the nasal field of view angle is greater than 60°, and the temporal field of view angle should be greater than 90°.

Images