A device and method for realizing eye tracking based on plastic optical fiber

Abstract

This application relates to a device and method for achieving eye tracking based on plastic optical fiber, belonging to the field of eye movement tracking technology. The device includes an acquisition module and a processing module. The acquisition module includes glasses and an optical fiber. The glasses include a frame. One end of the optical fiber is connected to the frame and the end is coated with an isolation layer. The optical fiber segment where the optical fiber contacts the frame is provided with a polishing area. When the tester wears the glasses, the polished surface of the polishing area faces the same direction as the tester's eye when looking straight ahead. The processing module includes an industrial camera and a server. The input end of the industrial camera is connected to the other end of the optical fiber, and the output end of the industrial camera is connected to the server. The industrial camera is used to generate a speckle signal. The server stores an eye movement recognition model, and the server obtains eye tracking results based on the speckle signal and the eye movement recognition model. This application has the effect of reducing the weight of the eye tracking equipment and improving the quality of eye tracking.

Description

Technical Field

[0001] The present application relates to the technical field of eye movement tracking, and in particular to a device and method for achieving eye tracking based on plastic optical fiber. Background Art

[0002] Eye tracking is the process of automatically locating the pupil center and gaze point. Humans primarily acquire external information through visual perception, and eye movements can intuitively reflect where and how long a person is looking. Eye tracking has applications in human-computer interaction, neuroscience research, and optometry. In the era of big data, eye tracking is also used to analyze consumer spending intentions and aid in medical diagnosis, such as diagnosing autism in infants and young children, Alzheimer's disease, and schizophrenia.

[0003] Eye tracking is mainly divided into two types: invasive tracking and non-invasive tracking. Among them, the typical invasive tracking is the electromagnetic scleral search coil, which is embedded in the contact lens and measures the horizontal and vertical movement of the eyeball through the signal generated by the electromagnetic induction principle. The movement of the eyeball will generate an induced voltage, and the eye movement is tracked by analyzing the electrical signal received by the system. This system has high accuracy and very low latency, but it is necessary to anesthetize the subject's eyes before the experiment, and then adsorb the device on the eyeball. This method is harmful to the subject's eyes. The application of invasive tracking also includes eye tracking devices such as the binocular Purkinje imaging system (DPI). Although this tracking device does not physically contact the eyes, it still requires a bite stick to stabilize the head, and the allowed field of view is very small, and the tracking accuracy is not high.

[0004] Non-invasive tracking includes video tracking method, iris edge method and pupil corneal reflection method. Among them, the video tracking method requires external equipment for support. Common methods include helmet-type, telemetry-type, and fixed-type, usually installing one or more cameras. When a camera is installed, an infrared light source is needed to cause the dark pupil effect. When the infrared light source is reflected on the eyeball, the pupil color is darker and the iris color is brighter. Through image analysis, the pupil can be separated to complete eye tracking. If multiple cameras are installed, only one camera is used to track the eyeball, and the remaining cameras are used to identify the position of the head to complete the calibration of the eye movement, which improves the accuracy of eye movement recognition to a certain extent.

[0005] The iris edge method illuminates the eye with infrared light. Two infrared photodiodes mounted near the eye receive infrared light reflected from the sclera and the iris edge. The amount of infrared light received varies with eye movement. When the eye moves to one side, the iris turns in that direction, reducing the amount of infrared light received by the photodiode on that side. Meanwhile, the sclera reflects more light on the other side, increasing the amount of infrared light received by the photodiode on that side. This differential signal enables contactless eye tracking. This method produces high horizontal accuracy, but suffers from low vertical accuracy, high interference, and significant head error.

[0006] The pupil-corneal reflection tracking method uses an eye camera to capture an eye image, then processes the image to determine the pupil center. The corneal reflection point is then used as the reference point for the relative position of the eye camera and the eyeball. The pupil center obtained from image processing can be used to derive the coordinates of the gaze vector, thereby determining the gaze point. After finding a good pupil-corneal reflection point, a calibration procedure is used to determine the mapping function between the vector formed by the pupil and corneal reflection points and the gaze point on the screen. By detecting the change in the pupil and corneal vectors, the point of gaze on the screen can be tracked in real time.

[0007] In summary, invasive tracking methods have been phased out of the market due to the potential damage they cause to the eyes. Eye tracking based on video or image recognition requires a camera mounted on a headset, increasing the weight of the device to achieve tracking results. However, this added weight can restrict the tester's natural eye movements, affecting accurate eye movement data collection. Furthermore, the added weight can be affected by the interaction of multiple components, leading to complex calibration and motion correction processes. This complex process increases test errors and makes eye tracking results unreliable. Summary of the Invention

[0008] In order to solve the problems of low quality of existing eye tracking results, complex eye tracking equipment, and heavy weight, the present application provides a device and method for eye tracking based on plastic optical fiber.

[0009] In a first aspect of the present application, a device for eye tracking based on plastic optical fiber is provided. The device includes an acquisition module and a processing module;

[0010] The acquisition module includes glasses and an optical fiber. The glasses include a frame. One end of the optical fiber is connected to the frame, and the end is coated with an isolation layer. A polishing area is provided on the optical fiber section where the optical fiber contacts the frame. When the tester wears the glasses, the polishing surface of the polishing area is oriented in the same direction as the tester's eyeball when looking straight ahead. The reflected light from the eyeball is reflected on the polishing area to obtain an incident signal.

[0011] The processing module includes an industrial camera and a server. The input end of the industrial camera is connected to the other end of the optical fiber, and the output end of the industrial camera is connected to the server. The industrial camera is used to generate a speckle signal, which is the speckle formed after the incident signal is transmitted through the optical fiber. The server stores an eye movement recognition model, and the server obtains eye movement tracking results by inputting the speckle signal into the eye movement recognition model.

[0012] By adopting the above technical solution, first, an optical fiber is configured on the eyeglass frame, and a polished area is provided on the surface of the optical fiber facing away from the eyeball. The polished area transmits the reflected light from the eyeball into the optical fiber. After the reflected light from the eyeball is transmitted by the optical fiber, speckle is formed in the industrial camera. The server then obtains the eye tracking results based on the speckle and the pre-stored eye movement recognition results. Therefore, in actual use, the tester only needs to wear the glasses as if they were wearing ordinary glasses, and their eye movements are not restricted. During the wearing process, the device of the present application can complete eye movement tracking based on the reflected light of the ambient light on the eyeball. The testing process is simple and easy to implement, which can improve the quality of the eye tracking results of the present application.

[0013] In one possible implementation, a plurality of parallel grooves are provided in the polishing area, the opening ends of the plurality of grooves are oriented in the same direction as the line of sight of the eyeball when looking straight ahead, the bottoms of the plurality of grooves are all at an angle of 45°, and the side walls of each of the plurality of grooves close to one end of the optical fiber coated with the isolation layer are all at an angle of 90°.

[0014] By adopting the above technical solution, multiple parallel grooves etched in the polishing area are used to collect the reflected light from the eyeball. The bottoms of the multiple grooves are set at 45 degrees, and the sidewalls of each of the multiple grooves near the end of the optical fiber coated with the isolation layer are all at 90 degrees. Therefore, the device of the present application can not only increase the coupling efficiency of the reflected light from the eye within the grooves, but also ensure that the reflected light is transmitted from the end of the optical fiber coated with the isolation layer to the end of the optical fiber connected to the industrial camera, thereby improving the accuracy of the obtained speckle signal.

[0015] In a possible implementation, the number of the grooves is three, and the three grooves are equidistantly distributed in the polishing area.

[0016] By adopting the above technical solution, the three grooves are evenly distributed in the polishing area, which can collect the reflected light from the eye in different directions. At the same time, it can also avoid too many factors affecting the distribution of speckle due to too many grooves, thereby reducing the difficulty of calibration.

[0017] In a possible implementation manner, a reflective layer is coated on the side walls of the plurality of grooves.

[0018] By adopting the above technical solution, a reflective layer can be coated on the side wall of the groove to reduce the coupling efficiency of the reflected light of the eyeball being refracted out of the optical fiber again, thereby improving the coupling efficiency of the reflected light of the eyeball.

[0019] In a possible implementation: the number of the optical fibers is at least two;

[0020] The frame includes a left frame and a right frame, each of the left frame and the right frame includes an upper edge and a lower edge, the upper edge is located above the eyeball, and the lower edge is located below the eyeball;

[0021] An optical fiber is connected to the upper edge and the lower edge of the left mirror frame respectively; and / or

[0022] An optical fiber is respectively connected to the upper edge and the lower edge of the right mirror frame.

[0023] By adopting the above technical solution, the present application can choose to track one or both eyes, and by providing optical fibers at the upper and lower edges, it avoids the situation where the reflected light from the eyeball cannot be collected due to the eyeball moving far away from the center of the frame, thereby ensuring the accuracy of the obtained speckle signal.

[0024] In a possible implementation, an infrared light source is further provided between the left frame and the right frame.

[0025] By adopting the above technical solution, in some environments with dim natural light, by turning on the infrared light source and enhancing the intensity of the reflected light of the eyeball, the purpose of collecting the reflected light of the eyeball in such an environment can be achieved, thereby improving the application scope of the device of the present application.

[0026] In a possible implementation, the optical fiber is a plastic optical fiber.

[0027] By adopting this technical solution, plastic optical fiber has a larger core diameter and numerical aperture than other types of optical fibers, such as quartz fiber, making it a multimode fiber. When reflected light from the eye travels through the plastic optical fiber, the various transmission modes interfere with each other, generating a speckle pattern of bright and dark spots at the output end of the fiber. This allows industrial cameras to capture and identify the speckle signal, providing hardware support for identifying and tracking eye movements.

[0028] In a second aspect of the present application, a method for implementing eye tracking based on plastic optical fiber is provided. The method comprises:

[0029] Acquire speckle signals;

[0030] The speckle signal is input into an eye movement recognition model to obtain an eye movement tracking result.

[0031] In a possible implementation, the eye tracking result is calculated using the following formula:

[0032] E(x,y)=O(ξ,η)×T(x,y;ξ,η),

[0033] Where T(x, y; ξ, η) is the transmission matrix of the optical fiber, O(ξ, η) and E(x, y) are the eye tracking results and speckle signals, respectively.

[0034] In one possible implementation, the eye movement recognition model is established by the following method:

[0035] collecting speckle signals corresponding to the industrial camera when the tester's eyeball moves along with the marker, and combining multiple speckle signals to form a test set;

[0036] Obtaining the coordinates of each movement of the marker to obtain an input signal, and combining multiple input signals to form a control set;

[0037] Establishing a nonlinear mapping relationship between each speckle signal in the test set and each input signal in the control set through a convolutional neural network to obtain a transmission matrix of the optical fiber;

[0038] An eye movement recognition model is obtained according to the transfer matrix.

[0039] In summary, this application includes at least one of the following beneficial technical effects:

[0040] First, an optical fiber is placed on the eyeglass frame, and a polished area is provided on the surface of the optical fiber facing away from the eyeball. The polished area transmits the reflected light from the eyeball into the optical fiber. After the reflected light from the eyeball is transmitted by the optical fiber, speckle is formed in the industrial camera. The server then obtains the eye tracking results based on the speckle and the pre-stored eye movement recognition results. Therefore, in actual use, the tester only needs to wear the glasses as if they were wearing ordinary glasses, and their eye movements are not restricted. During the wearing process, the device of the present application can complete eye movement tracking based on the reflected light of the ambient light on the eyeball. The testing process is simple and easy to implement, which can improve the quality of the eye tracking results of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a schematic diagram of the structure of a device for achieving eye tracking based on plastic optical fiber according to an embodiment of the present application.

[0042] Figure 2 Schematic diagram of the polishing area on the optical fiber in the embodiment of the device of the present application.

[0043] Figure 3 yes Figure 2 Schematic diagram of how multiple grooves in the polished area transmit the reflected light from the eye into the optical fiber.

[0044] Figure 4 This is a schematic diagram of the corresponding relationship between the frame and the eyeball in the embodiment of the present application.

[0045] Figure 5 This is a flow chart of a method for realizing eye tracking based on plastic optical fiber according to an embodiment of the present application.

[0046] Figure 6 This is a system block diagram of an embodiment of the present application that implements eye tracking based on plastic optical fiber.

[0047] Explanation of the accompanying symbols: 1. glasses; 2. eyeball; 3. optical fiber; 31. polishing area; 4. industrial camera; 5. server; 51. data acquisition unit; 52. data generation unit. DETAILED DESCRIPTION

[0048] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0049] The embodiments of the present application are described in further detail below with reference to the accompanying drawings.

[0050] Because invasive eye tracking methods currently cause damage to the eyes, they are no longer applicable in the current eye tracking market. Non-invasive tracking methods use tracking devices that integrate multiple components, resulting in large size and high power consumption. Furthermore, prolonged use can cause significant discomfort to testers, reducing their experience. Therefore, this application proposes a device for eye tracking based on plastic optical fiber.

[0051] like Figure 1 As shown, the device for realizing eye tracking based on plastic optical fiber includes an acquisition module and a processing module, which are connected. The acquisition module is used to collect the eye movement data of the tester, and the processing module is used to obtain the eye movement tracking results of the tester based on the eye movement data and a pre-set eye movement recognition model.

[0052] Specifically, the acquisition module includes glasses 1 and optical fibers 3. The glasses 1 include a frame and temples. The frame is divided into a left frame and a right frame, and the temples are divided into a left temple and a right temple. The left frame is connected to the left temple, and the right frame is connected to the right temple. The left frame is also connected to the right frame via a bridge. In a specific example, after the tester wears the glasses 1, the edge above the tester's eyeball 2 is used as the upper edge of the frame, and the edge below the tester's eyeball 2 is used as the lower edge of the frame. Therefore, the left frame includes an upper edge and a lower edge, and the right frame also includes an upper edge and a lower edge.

[0053] In a specific example, the optical fiber 3 can be laid out on the left or right frame, and the eye movement data of the tester's single eye can be collected through the optical fiber 3, that is, the movement of the tester's eyeball 2 can be known by collecting the movement of the single eye. Of course, the optical fiber 3 can also be laid out on both the left and right frames at the same time, and the eye movement data of the tester's two eyes can be collected through the optical fiber 3 to know the movement of the tester's eyeball 2. In this example, whether the eye movement data is obtained in a monocular collection mode or a binocular collection mode, the arrangement of the optical fiber 3 on the frame is the same. Therefore, in order to facilitate the explanation of the arrangement of the optical fiber 3, the monocular collection mode in which the optical fiber 3 is laid out on the right frame is taken as an example below:

[0054] An optical fiber 3 is connected to the upper edge of the right frame, and its port is close to the bridge between the left and right frames. The port is coated with an isolation layer made of ink to prevent external stray light from entering the optical fiber 3. The other end of the optical fiber 3 is connected to the processing module;

[0055] Another optical fiber 3 is connected to the lower edge of the right frame, and its port is close to the bridge set between the left and right frames. The port is also coated with an isolation layer. The other end of the optical fiber 3 is merged with the optical fiber 3 set at the upper edge at the processing module. Specifically, the ports of the two optical fibers 3 are glued together using glue, and then the common port obtained by bonding is connected to the input end of the processing module.

[0056] It should be noted that the optical fibers 3 used in this application are all plastic optical fibers, which have a larger core diameter and a larger numerical aperture than other types of optical fibers, such as quartz optical fibers, and are multimode optical fibers. For ease of distinction, the optical fiber 3 connected to the upper edge of the right frame is referred to as the first optical fiber 3, and the optical fiber 3 connected to the lower edge of the right frame is referred to as the second optical fiber 3. Polishing areas 31 are provided on both the first optical fiber 3 and the second optical fiber 3. In this example, the polishing areas 31 of the first optical fiber 3 and the second optical fiber 3 are symmetrical with respect to the line connecting the corner of the tester's eye to the end of the eye. Therefore, in order to specifically illustrate the polishing areas 31 on the first optical fiber 3 and the second optical fiber 3, the polishing area 31 on the first optical fiber 3 is taken as an example below:

[0057] like Figure 2As shown, the polishing area 31 is located on the fiber segment where the first optical fiber 3 contacts the upper edge of the right frame. The polishing surface of the polishing area 31 is oriented in the same direction as the line of sight of the tester's eyeball 2 when looking straight ahead. Specifically, an angle grinder is used to etch multiple parallel grooves on the polishing surface in the polishing area 31, each with a bottom angle of 45° and a sidewall angle of 90° near the end of the optical fiber 3 coated with the isolation layer. The opening ends of the multiple grooves are oriented in the same direction as the line of sight of the eyeball 2 when looking straight ahead. In a specific example, there are three grooves, and the three grooves are equidistantly distributed in the polishing area 31, that is, one groove is located in the center of the polishing area 31, and the other two grooves are symmetrically located on both sides of the groove at the center. At the same time, the line connecting the three grooves is parallel to the line connecting the corner to the end of the tester's eye. It should be noted that, in actual use, since the corners and tails of the tester's eyes are not necessarily on the same horizontal plane, the line connecting the three grooves is not necessarily parallel to the line connecting the corners of the tester's eyes to the tails of the eyes, but as long as it is guaranteed to be within a preset fluctuation range, it can ensure that the optical fiber 3 on the upper edge and the optical fiber 3 on the lower edge work together to collect the reflected light of the eyeball 2 in all directions.

[0058] In this example, the three grooves in the polishing area 31 are used to transmit the reflected light of the eyeball 2 to the optical fiber 3, so as to transmit the reflected light of the eyeball 2 from the end coated with the isolation layer to the end connected to the processing module. Figure 3 For example, point A is the position of the eyeball 2. The light reflected from point A is reflected by the side wall of the optical fiber 3, and then reflected again by the side wall of the groove, and then moves along the optical fiber 3 toward the end connected to the processing module.

[0059] It should be noted that in other examples, optical fibers 3 may also be provided on the left and right edges between the upper and lower edges. However, since the polishing area 31 of the upper edge and the polishing area 31 of the lower edge of the present application have already covered the entire eyeball 2, as shown in FIG. Figure 4 As shown, the eyeball 2 is located between the polishing area 31 of the optical fiber 3 on the upper edge and the polishing area 31 of the optical fiber 3 on the lower edge. The lengths of the two polishing areas 31 are both greater than the diameter of the eyeball 2, so setting the optical fiber 3 on the upper edge and the optical fiber 3 on the lower edge is sufficient to meet the needs of eye tracking.

[0060] The processing module includes an industrial camera 4 and a server 5. Among them, the industrial camera 4 is also commonly known as an industrial camera. Compared with traditional cameras, it has higher image stability, high transmission capacity and high anti-interference ability. Most of the industrial cameras 4 on the market are cameras based on charge-coupled devices (CCD) or complementary metal oxide semiconductor (CMOS) chips. In this application, the industrial camera 4 adopts a CCD-based industrial camera 4. The industrial camera 4 and the charge-coupled device are integrated into one design to reduce the space volume of the processing module and avoid setting up a connection line between the industrial camera 4 and the charge-coupled device, which makes the light transmission path longer and the interference factor increases, that is, to ensure the accuracy of the speckle generated in the industrial camera 4.

[0061] A charge-coupled device (CCD) is a solid-state electronic device used to detect light. It uses a clock pulse voltage to generate and control changes in a semiconductor potential well, enabling the storage and transmission of charge information. In this example, a CCD is used to combine two optical fibers (3) connected to the same frame. This involves gluing the ends of the first and second optical fibers (remote from the bridge) together. The resulting common port is then connected to the CCD input. Simultaneously, the common port of the two optical fibers (3) captures the light reflected from the eyeball and images it on an industrial camera (4), producing speckle.

[0062] Furthermore, since this application measures the speckle pattern formed in the industrial camera 4 by the reflected light from the tester's eyeball 2 under natural light, when the tester is in a dimly lit environment, the reflected light from the eyeball 2 will be relatively weak, making the speckle pattern formed in the industrial camera 4 less noticeable. To this end, a low-power infrared light source (not shown in the accompanying drawings) is provided between the left and right frames. When the tester is in a dimly lit environment, the infrared light source provided between the left and right frames enhances the ambient light intensity, allowing a single infrared light source to meet the lighting requirements of both monocular and binocular acquisition modes, while also conserving space in the acquisition module.

[0063] This application uses plastic optical fiber to collect the reflected light from the eyeball 2 and transmit it to the industrial camera 4 for recording and identification, which can reduce the weight of VR or other equipment that needs to record the direction of eye movement. In addition, by receiving the reflected light of natural light on the eyeball 2 to complete eye tracking, in many scenarios, there is no need to use an additional infrared light source to illuminate the eyeball 2 to generate reflected light. This invention can reduce the power consumption of the acquisition module. In addition, in actual use, the tester only needs to wear glasses 1 as usual, and his eye movement is not restricted. The tester has a larger range of activities, which makes the accuracy of the speckle formed in the industrial camera 4 higher.

[0064] Figure 5 A flow chart of a method for realizing eye tracking based on plastic optical fiber according to an embodiment of the present application is shown. Figure 1 The main process of the method for realizing eye tracking based on plastic optical fiber is described as follows.

[0065] Step S10: Acquire speckle signal.

[0066] As can be seen from the device embodiment, the speckle signal is the speckle formed in the industrial camera 4 after the reflected light from the eyeball 2 is transmitted through the optical fiber 3. The industrial camera 4 is connected to the server 5. Therefore, after the industrial camera 4 obtains the speckle signal, it transmits the speckle signal to the server 5 in real time. In other words, the server 5 obtains the speckle signal.

[0067] Step S20: inputting the speckle signal into the eye movement recognition model to obtain the eye movement tracking result.

[0068] First, an eye movement recognition model is established in advance. The specific establishment process is shown in steps S11 to S13:

[0069] Step S11: A display screen is provided in the tester's environment, displaying markers at different locations at different times, such as a moving circle or other moving pattern. The display screen is connected to a server 5, and the coordinates of the moving markers on the display screen are transmitted to the server 5 in real time. Alternatively, the server 5 intelligently controls the coordinates of the markers on the display screen. Therefore, the server 5 has pre-planned the markers' movement routes and thus knows the coordinates of the markers.

[0070] Step S12: Collect the speckle patterns output by the industrial camera 4 when the eyeball 2 moves to different positions, and generate a test set based on these multiple speckle patterns. Simultaneously, as the tester observes the markers that appear one after another on the display screen, the server 5 also acquires the coordinates of the markers in real time. Each coordinate of the marker movement serves as an input signal, and a control set is generated based on the coordinates corresponding to these multiple marker movements. This control set, therefore, contains multiple input signals.

[0071] Step S13: When light is transmitted in the plastic optical fiber, various transmission modes cause intermodal interference in the optical fiber 3, generating a speckle pattern consisting of bright spots and dark spots at the output end of the optical fiber 3. This speckle pattern is also called a speckle signal. In this example, the speckle signal can be expressed as:

[0072]

[0073] Where α and φ represent the amplitude and phase of the transmission mode of the optical fiber 3, respectively; x and y are the coordinates of the speckle signal on the plane where the eye 2 is looking; M is the total number of transmission modes; and m and n are the different transmission modes in the optical fiber 3, respectively.

[0074] In other words, server 5 simultaneously records the speckle shape in the speckle pattern and the coordinates of the marker observed by the tester. By training a convolutional neural network, it can extract the features of these speckle patterns and establish a nonlinear mapping relationship between them and the coordinates of the marker. This nonlinear mapping relationship is also called the transmission matrix during transmission through optical fiber 3. Server 5 saves the trained network model as an eye movement recognition model, thus completing the calibration process.

[0075] It should be noted that the calibration accuracy of the eye movement recognition model is proportional to the density of the markers observed by the tester, so the density of the markers can be increased appropriately during calibration.

[0076] Therefore, based on the established eye movement recognition model, this application inputs the speckle signal into the eye movement recognition model, which processes and recognizes the features of the speckle pattern. Then, based on the nonlinear mapping relationship between the features of the speckle pattern and the coordinates of the marker, the tester's eye tracking results are obtained. Specifically, the calculation formula for calculating the eye tracking results is as follows:

[0077] E(x,y)=O(ξ,η)×T(x,y;ξ,η),

[0078] Where T(x, y; ξ, η) is the transmission matrix of optical fiber 3, O(ξ, η) and E(x, y) are the eye tracking results and speckle signal, respectively. In other words, based on the pre-established eye movement recognition model, eye movement (i.e., the movement of the marker) can be quickly calculated after acquiring the speckle signal.

[0079] In summary, the implementation principle of the method for achieving eye tracking based on plastic optical fiber in the embodiment of the present application is as follows: This application performs eye tracking based on the reflection of ambient light on the eye 2. When the tester is moving during the test, the reflected light from the eye 2 may change with the relative position of the person and the light source. Based on the pre-established nonlinear mapping relationship between the characteristics of the speckle pattern and the coordinates of the marker, the coordinates of the marker can be quickly deduced from the speckle pattern, thereby achieving the purpose of eye tracking. In addition, in actual use, the tester only needs to wear the glasses 1 as usual, and their eye movements are not restricted. The collected speckle pattern is more accurate, which improves the quality of the eye tracking results.

[0080] Figure 6 The block diagram of a system for realizing eye tracking based on plastic optical fiber according to an embodiment of the present application is shown. Figure 1The server 5 includes a data acquisition unit 51 and a data generation unit 52.

[0081] The data acquisition unit 51 is configured to acquire a speckle signal.

[0082] The data generating unit 52 is configured to input the speckle signal into the eye movement recognition model to obtain an eye movement tracking result.

[0083] The units described in the embodiments of this application may be implemented in software or hardware. The units described may also be provided in a processor. For example, it may be described as follows: a processor includes a data acquisition unit 51 and a data generation unit 52. The names of these units do not, in some cases, limit the units themselves. For example, the data acquisition unit 51 may also be described as a "unit for acquiring speckle signals."

[0084] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the described module can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0085] The present application also provides a computer-readable storage medium, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, among other media capable of storing program code. The computer-readable storage medium stores a computer program capable of being loaded by a processor and executing the aforementioned method for implementing eye tracking based on plastic optical fiber.

[0086] The above description is merely a preferred embodiment of the present application and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this application is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but also encompasses other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the aforementioned disclosed concepts. For example, a technical solution formed by replacing the above-mentioned features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A device for eye tracking based on plastic optical fiber, characterized by: Includes acquisition module and processing module; The acquisition module comprises glasses (1) and an optical fiber (3), wherein the glasses (1) comprise a frame, one end of the optical fiber (3) is connected to the frame and the end is coated with an isolation layer, and a polishing area (31) is provided on the optical fiber section of the optical fiber (3) in contact with the frame, and when the tester wears the glasses (1), the direction of the polishing surface of the polishing area (31) is consistent with the direction of sight of the tester's eyeball (2) when looking straight ahead, and the reflected light of the eyeball (2) is reflected in the polishing area (31) to obtain an incident signal; The processing module includes an industrial camera (4) and a server (5), wherein the input end of the industrial camera (4) is connected to the other end of the optical fiber (3), and the output end of the industrial camera (4) is connected to the server (5), and the industrial camera (4) is used to generate a speckle signal, wherein the speckle signal is a speckle formed after the incident signal is transmitted through the optical fiber (3); the server (5) stores an eye movement recognition model, and the server (5) obtains an eye movement tracking result by inputting the speckle signal into the eye movement recognition model; The optical fiber (3) is a plastic optical fiber.

2. The device for realizing eye tracking based on plastic optical fiber according to claim 1, characterized in that: A plurality of parallel grooves are provided in the polishing area (31), the opening ends of the plurality of grooves are oriented in the same direction as the line of sight of the eyeball (2) when looking horizontally, the bottoms of the plurality of grooves are all at an angle of 45 degrees, and the sidewalls of each of the plurality of grooves close to one end of the optical fiber (3) coated with the isolation layer are all at an angle of 90 degrees.

3. The device for realizing eye tracking based on plastic optical fiber according to claim 2, characterized in that: The number of the grooves is three, and the three grooves are equidistantly distributed in the polishing area (31).

4. The device for realizing eye tracking based on plastic optical fiber according to claim 2, characterized in that: The side walls of the plurality of grooves are coated with a reflective layer.

5. The device for realizing eye tracking based on plastic optical fiber according to claim 1, characterized in that: The number of the optical fibers (3) is at least two; The frame comprises a left frame and a right frame, and both the left frame and the right frame comprise an upper edge and a lower edge, the upper edge being located above the eyeball (2), and the lower edge being located below the eyeball (2); An optical fiber (3) is connected to the upper edge and the lower edge of the left mirror frame respectively; and / or An optical fiber (3) is connected to the upper edge and the lower edge of the right mirror frame respectively.

6. The device for realizing eye tracking based on plastic optical fiber according to claim 5, characterized in that: It also includes an infrared light irradiation source arranged between the left mirror frame and the right mirror frame.

7. A method for eye tracking based on plastic optical fiber, characterized in that: include: Acquire speckle signals; Inputting the speckle signal into an eye movement recognition model to obtain an eye movement tracking result; The eye tracking results are calculated using the following formula: E(x, y)=O(ξ, η)×T(x, y; ξ, η), Where T(x, y; ξ, η) is the transmission matrix of the optical fiber (3), O(ξ, η) and E(x, y) are the eye tracking results and speckle signals, respectively; The eye movement recognition model is established by the following method: collecting speckle signals corresponding to the eyeballs (2) of the tester and the movement of the marker by the industrial camera (4), and combining a plurality of the speckle signals to form a test set; Obtaining the coordinates of each movement of the marker to obtain an input signal, and combining multiple input signals to form a control set; Establishing a nonlinear mapping relationship between each speckle signal in the test set and each input signal in the control set through a convolutional neural network to obtain a transmission matrix of the optical fiber (3); An eye movement recognition model is obtained according to the transfer matrix.

Images