Myopia prevention and control method based on wearable device, wearable device and storage medium

Abstract

The invention discloses a myopia prevention and control method based on a wearable device, the wearable device and a storage medium, and relates to the technical field of wearable device.The method comprises the steps that a first image representing user eye information is collected, and a second image representing a user cornea area in the first image is extracted; performing spherical distortion removal processing on the second image to obtain a target cornea reflection image; multi-dimensional dynamic analysis is carried out on the target cornea reflection image to obtain multi-dimensional evaluation information, and the multi-dimensional dynamic analysis comprises target object recognition, distance speculation and illumination detection; in response to the situation that the multi-dimensional evaluation information does not meet preset myopia prevention and control conditions, it is determined that the myopia risk exists, preset prompt information for prompting the user that the myopia risk exists is output, and the preset myopia prevention and control conditions comprise at least one of the fixation time condition, the fixation distance condition and the illumination intensity condition. According to the method, the hardware cost of the wearable equipment is not increased, and meanwhile, accurate myopia prevention and control are realized.

Description

Technical Field

[0001] This application relates to the field of wearable device technology, and in particular to myopia prevention and control methods, wearable devices, and storage media based on wearable devices. Background Technology

[0002] In recent years, with the widespread use of electronic products and the increasing academic burden, the global incidence of myopia has shown a trend of high prevalence and younger age of onset, making the vision health of children and adolescents a serious public health challenge. Medical research shows that poor eye-use behavior is a core environmental factor leading to the occurrence and development of myopia. Therefore, accurate monitoring and intervention of users' daily eye behavior is a key link in myopia prevention and control. However, current methods of myopia prevention and control require adding other devices to wearable devices (such as AR glasses), such as adding distance sensors to determine the current viewing distance. This reliance solely on viewing distance for myopia prevention and control fails to achieve precise control, and the risk of myopia still exists due to users' poor habits (such as viewing electronic products in dimly lit environments).

[0003] Therefore, how to achieve accurate myopia prevention and control without increasing the hardware cost of wearable devices has become an urgent problem to be solved.

[0004] The above content is only used to help understand the technical solution of this application and does not represent an admission that the above content is prior art. Summary of the Invention

[0005] The main objective of this application is to provide a myopia prevention and control method, a wearable device, and a storage medium based on wearable devices, aiming to solve the technical problem of how to achieve accurate myopia prevention and control without increasing the hardware cost of wearable devices.

[0006] To achieve the above objectives, this application proposes a myopia prevention and control method based on a wearable device. The method is applied to a wearable device, which includes at least one camera, and includes the following steps: Acquire a first image representing the user's eye information, and extract a second image representing the user's corneal region from the first image; The second image is subjected to spherical distortion correction to obtain the target corneal reflection image; Multi-dimensional dynamic analysis of the target corneal reflection image is performed to obtain multi-dimensional evaluation information, including target object recognition, distance estimation, and illumination detection. In response to the fact that the multi-dimensional assessment information does not meet the preset myopia prevention and control conditions, the system determines that there is a risk of myopia and outputs a preset prompt message to the user indicating that there is a risk of myopia. The preset myopia prevention and control conditions include at least one of the following: fixation time condition, fixation distance condition, and light intensity condition.

[0007] Optionally, the step of performing spherical distortion correction on the second image to obtain the target corneal reflection image includes: Construct an optical model of the eyeball corresponding to the user wearing the wearable device, and determine the spherical coordinate system corresponding to the optical model of the eyeball; Determine the mapping relationship between the spherical coordinate system and the Cartesian coordinate system corresponding to the second image; Based on the mapping relationship, each pixel in the second image is mapped into a ray emanating from the shooting end, and a corneal reflection image is constructed based on the intersection of each ray emanating from the eye's optical model. The corneal reflection image is enhanced with detail processing to obtain the target corneal reflection image.

[0008] Optionally, the step of performing multi-dimensional dynamic analysis on the target corneal reflection image to obtain multi-dimensional evaluation information includes: In response to multi-dimensional dynamic analysis including target object recognition, high-frequency feature recognition is performed on the target corneal reflection image, and the existence of a first image region representing a rectangular contour is detected in the target corneal reflection image; In response to the presence of high-frequency features in the target corneal reflection image and the existence of a first image region, a second image region other than the first image region in the target corneal reflection image is determined, and it is detected whether the image parameters of the first image region and the image parameters of the second image region match, and whether the color temperature information of the ROI region in the target corneal reflection image is less than a preset color temperature threshold. If the image parameters of the first image region and the image parameters of the second image region do not match, and the color temperature information of the ROI region is less than the preset color temperature threshold, then the multi-dimensional evaluation information is determined to include the target object that the user is looking at.

[0009] Optionally, distance estimation includes coarse distance measurement, and the shooting end includes a left-eye camera and a right-eye camera. The steps for performing multi-dimensional dynamic analysis on the target corneal reflectance image to obtain multi-dimensional evaluation information include: In response to multi-dimensional dynamic analysis, including coarse distance measurement, the left and right eye images corresponding to the target corneal reflection image are determined, wherein the left and right eye images are images captured by the left and right eye cameras respectively at the same time step; The binocular eye gaze angle is calculated based on the left and right eye images. The first distance between the user and the target object is determined based on the binocular eye gaze angle. Multi-dimensional evaluation information, including the first distance, is also determined. The larger the binocular eye gaze angle, the smaller the first distance.

[0010] Optionally, distance estimation includes precise distance measurement. The steps for performing multi-dimensional dynamic analysis on the target corneal reflectance image to obtain multi-dimensional evaluation information include: In response to multi-dimensional dynamic analysis, including distance precision measurement, the system identifies a third image region representing the target object in the corneal reflection image and determines the pixel size parameter information of the third image region. The actual size parameters of the target object are determined. Based on the comparison between the pixel size parameters and the actual size parameters, the second distance between the user and the target object is determined, and multi-dimensional evaluation information including the second distance is determined.

[0011] Optionally, the step of performing multi-dimensional dynamic analysis on the target corneal reflection image to obtain multi-dimensional evaluation information further includes: In response to multi-dimensional dynamic analysis including illumination detection, the system determines the first capture time of the first image corresponding to the target corneal reflection image, acquires the illumination intensity of the environment in which the wearable device is located at the first capture time, and determines multi-dimensional evaluation information including illumination intensity.

[0012] Optionally, after the step of performing multi-dimensional dynamic analysis on the target corneal reflection image to obtain multi-dimensional evaluation information, at least one of the following is included: In response to multi-dimensional evaluation information, including whether the user is looking at the target object, if the continuous gaze time of the user looking at the target object is greater than the preset gaze time, it is determined that the multi-dimensional evaluation information does not meet the gaze time condition. In response to multi-dimensional evaluation information including a first distance and / or a second distance, if the first distance and / or the second distance are less than a preset distance threshold, it is determined that the multi-dimensional evaluation information does not meet the gaze distance condition; In response to multi-dimensional evaluation information including light intensity, if the light intensity is less than a preset light intensity threshold, it is determined that the multi-dimensional evaluation information does not meet the light intensity condition.

[0013] Optionally, the step of extracting a second image representing the user's corneal region from the first image includes: The image region covering the corneal area in the first image is determined, and the image region covering the corneal area is segmented to obtain the corneal area image. The light spot information in the corneal area image is filtered out to obtain the second image.

[0014] In addition, to achieve the above objectives, this application also proposes a wearable device, which includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, the computer program being configured to implement the steps of the myopia prevention and control method based on the wearable device as described above.

[0015] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the myopia prevention and control method based on wearable devices as described above.

[0016] In this application, myopia prevention and control is achieved by utilizing the existing camera on the wearable device itself. Specifically, a first image representing the user's eye information is acquired, and a second image representing the user's corneal region is extracted from it. Spherical distortion correction processing is then performed to obtain a target corneal reflection image. This target corneal reflection image is then subjected to multi-dimensional dynamic analysis (such as target object recognition, distance estimation, and illumination detection) to obtain multi-dimensional evaluation information. When the multi-dimensional evaluation information does not meet preset myopia prevention and control conditions (such as fixation time, fixation distance, and illumination conditions at least one), a myopia risk is determined, and a preset warning message indicating the myopia risk is output to the user. This allows for accurate prediction using the target corneal reflection image without increasing additional hardware costs. A comprehensive evaluation is conducted from multiple dimensions (such as fixation time, fixation distance, and illumination conditions) to improve the accuracy of myopia prevention and control for users. Therefore, accurate myopia prevention and control can be achieved without increasing the hardware cost of the wearable device. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the architecture of a wearable device in an embodiment of this application; Figure 2 This application provides a flowchart illustrating the first embodiment of a myopia prevention method based on wearable devices; Figure 3 A flowchart is provided in the second embodiment of the myopia prevention method based on wearable devices in this application; Figure 4 This is another flowchart illustrating the myopia prevention method based on wearable devices in this application; Figure 5 This is a schematic diagram of the hardware operating environment involved in the myopia prevention and control method based on wearable devices in the embodiments of this application.

[0020] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0021] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0022] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0023] Optionally, the execution subject of the myopia prevention and control method based on wearable devices in the embodiments of this application can be a wearable device (which can be an AR device, a VR device, such as smart glasses, etc.).

[0024] Optionally, this application uses smart glasses as an example of a wearable device for illustration.

[0025] Optionally, when carrying out myopia prevention and control, various sensors and other hardware can be integrated into smart glasses to achieve myopia prevention and control for users, such as in the following ways.

[0026] Method 1: Computer vision recognition based on a front-facing camera. A front-facing camera can be installed on the bridge of the nose or the outside of the frame of the smart glasses to directly capture the field of vision in front of the wearer (i.e., the user wearing the smart glasses). Image recognition algorithms are used to determine whether the wearer is reading or looking at a screen, and distance is estimated by combining the size of objects. However, this method has certain drawbacks, such as privacy violations. In sensitive environments such as schools, changing rooms, and confidential places, smart glasses with cameras are often prohibited from use. Furthermore, this method requires real-time processing of high-definition video streams, which places extremely high demands on chip computing power, resulting in high power consumption and severe heat generation in the smart glasses, making it difficult to meet the needs of wearers for all-day wear.

[0027] Method 2, based on physical measurement using distance sensors. Smart glasses can use sensors such as infrared time-of-flight sensors, ultrasonic sensors, or lidar to measure the distance between the smart glasses and objects in front of the user. However, these sensors can only obtain distance data from the center point and have difficulty distinguishing object attributes (for example, they cannot distinguish whether the object in front of the user is a wall or a glowing mobile phone screen). Furthermore, distance sensors are expensive and require a large volume and structural space in the smart glasses, resulting in bulky and heavy smart glasses that do not meet the requirements of children and teenagers for lightweight and imperceptible glasses.

[0028] Method 3, eye-tracking method, uses an infrared camera inside the smart glasses to capture eye images and employs the central pupillary corneal reflection (PCCR) method for calculation. In this method, the corneal reflection image (such as the first Pkinje image (P1)) is only treated as a bright feature point to construct the gaze vector to calculate the gaze point coordinates. However, this method has the following drawbacks: a) Low information utilization: Since Method 3 treats P1 as a simple geometric coordinate point, it ignores the fact that P1 itself is a mirror image formed by ambient light on the convex surface of the cornea. It does not utilize the image content of P1 or the second Pkinje image (P2) (such as the reflected content of the mobile phone screen, distorted rectangular outline) to infer environmental information; b) Lack of depth perception: The PCCR algorithm solves the gaze direction problem in a two-dimensional plane (i.e., "where is being looked in this plane"), and it is difficult to directly calculate the absolute distance of the gaze target (i.e., "how far away is the thing being looked from the eyes") solely from eye-tracking data; c) Hardware dependence: In order to obtain environmental information, it is necessary to cooperate with a front-facing camera to perform "viewpoint-scene" mapping.

[0029] Alternatively, methods 1 through 3 all struggle to find a balance between accurate environmental perception and privacy protection / low cost / lightweight design. They cannot achieve the goal of extracting information about the wearer's eye distance, screen usage status, and ambient light from the collected corneal reflection images (P1, P2) using only the existing eye-tracking hardware on the inside of the glasses without adding a front-facing camera and a distance sensor.

[0030] Therefore, to avoid the above-mentioned defects, in the embodiments of this application, it is possible to accurately reverse calculate the wearer's eye distance, electronic screen usage status and ambient light intensity by using only corneal reflection images collected by a wearable device, such as a camera inside smart glasses, without adding additional hardware.

[0031] Optionally, this embodiment can omit the "front-facing camera" required in method 1, and indirectly obtain environmental information by utilizing the physical characteristics of corneal reflection. This eliminates the possibility of filming the external environment and irrelevant personnel from the hardware perspective, completely resolving privacy concerns in scenarios such as schools and homes, achieving comprehensive privacy protection. Furthermore, it eliminates the need for additional LiDAR, ToF ranging sensors, or ambient light sensors; multidimensional perception can be achieved by reusing existing eye-tracking cameras through algorithms. This not only significantly reduces hardware costs but also decreases device size and power consumption, making the glasses lighter and more suitable for children and adolescents to wear all day, thus achieving low cost and lightweight design. Moreover, this embodiment can calculate three key indicators—"viewing distance," "electronic screen usage status," and "ambient light intensity"—while determining the gaze point coordinates, filling the blind spots in myopia prevention data collection found in method 3, providing precise evidence for personalized vision health intervention, and achieving precise and comprehensive utilization of data dimensions. In other words, it can achieve precise myopia prevention without increasing the hardware cost of wearable devices.

[0032] Optionally, the hardware architecture of the wearable device in the embodiments of this application can refer to the following: Figure 1 As shown, when the wearable device is smart glasses, it can be integrated into the glasses frame. The wearable device may include a central processing unit, a visual acquisition module connected to the central processing unit, a posture sensing module, a photoresistor, a storage module, an interaction and communication module, and a power management module, and the power management module can power the central processing unit.

[0033] Optionally, the visual acquisition module includes a high-resolution endoscopic camera module located on the side or lower edge of the nose pad of the frame, with the lens focused on the wearer's eye area. The camera requirements are higher than those for eye-tracking methods, requiring macro shooting capabilities and high dynamic range (HDR) to ensure clear capture of the pupil edge while simultaneously imaging the faint environmental reflections on the corneal surface (i.e., the P1 image).

[0034] Optionally, the high-resolution endoscopic camera module may include two imaging ends (such as a left-eye endoscopic camera and a right-eye endoscopic camera), and these two cameras are connected to a power management module to power the two cameras. The left-eye endoscopic camera and the right-eye endoscopic camera can send the acquired image data to the central processing unit.

[0035] Optionally, the vision acquisition module also includes a near-infrared light source module, symmetrically distributed on the inner side of the frame, emitting invisible near-infrared light with wavelengths of 850nm or 940nm towards the wearer's eyes. This light source is mainly used to generate a clear corneal reflective spot (Glint) to assist eye-tracking positioning, and also serves as a supplementary light to illuminate eye features in low-light environments. Therefore, the central processing unit can send control commands to control the near-infrared light source module.

[0036] Optionally, the central processing unit can integrate a low-power chip for ISP (Image Signal Processing) and NPU (Neural Processing Unit) to handle image acquisition, distortion correction algorithm operation, and scene analysis.

[0037] Optionally, the attitude sensing unit includes an IMU (Inertial Measurement Unit) that transmits the detected attitude data to the central processing unit. A GM (Glass Measuring Unit) can be housed within the photoresistor to transmit the detected ambient light intensity (i.e., illumination intensity) to the central processing unit. The interaction and communication module includes a Bluetooth / wireless communication module, enabling the central processing unit to interact with external terminal devices via data transmission. The interaction and communication module also includes a vibration motor / feedback module, to which the central processing unit can send drive signals to generate vibration or feedback signals. The central processing unit can also perform read and write operations in the storage module.

[0038] Optionally, the multimodal sensor array (such as an IMU inertial measurement unit and a photoresistor) in the wearable device can be used to determine the user's head posture (such as looking down or looking up) to assist in determining reading posture and obtaining the current ambient light intensity.

[0039] Optionally, the myopia prevention and control method based on wearable devices in this application embodiment can be executed by using the central processing unit of a wearable device in combination with other hardware sensors.

[0040] Based on this, the embodiments of this application provide a myopia prevention and control method based on wearable devices, referring to... Figure 2 , Figure 2 This is a flowchart illustrating the first embodiment of the myopia prevention and control method based on wearable devices in this application.

[0041] In this embodiment, the myopia prevention and control method based on wearable devices is applied to wearable devices, which include at least one camera end, and the myopia prevention and control method based on wearable devices includes steps S10 to S40.

[0042] Step S10: Acquire a first image representing the user's eye information, and extract a second image representing the user's corneal region from the first image; Optionally, the wearable device is in a wearing state by the user and is in normal working condition.

[0043] Optionally, after detecting that the wearable device is in a wearing state and the myopia prevention function is enabled, the wearable device can take a picture of the user's eyes through at least one built-in camera to obtain an image of the user's eyes and use it as the first image representing the user's eye information.

[0044] Optionally, the user's eyes can be photographed using the left-eye endoscopic camera of the wearable device to obtain a left-eye image, and the left-eye image can be used as the first image; and / or, the user's eyes can be photographed using the right-eye endoscopic camera of the wearable device to obtain a right-eye image, and the right-eye image can be used as the first image.

[0045] Optionally, the same processing can be performed on the left-eye image and the right-eye image, and steps S10-S20 can be executed on both. In this embodiment, only a first image is used as an example for illustration.

[0046] Optionally, after acquiring the first image, a second image representing the user's corneal region can be identified and extracted from the first image. For example, an image recognition model can be used for identification and extraction, or the central pupillary corneal reflectance method can be used for processing; there are no restrictions on this.

[0047] Optionally, step S10, which involves extracting a second image representing the user's corneal region from the first image, further includes step a10.

[0048] Step a10 determines the image region covering the corneal region in the first image, segments the image region covering the corneal region to obtain the corneal region image, filters out the spot information in the corneal region image to obtain the second image.

[0049] Alternatively, the pupil-center corneal reflectance (PCCR) method can be used to locate the pupil and corneal regions in the first image based on Har features or a deep learning-based U-Net network, and use them as the image region covering the corneal region.

[0050] Optionally, Haar features are rectangular features that extract features by calculating the difference between pixels within a rectangular region of the image. These features can represent structures such as edges, lines, and centers. Therefore, the pupil position in the first image can be determined using Haar features, and then the cornea can be searched around the pupil position to obtain the corneal region. The corneal region is a rectangular or elliptical area that extends outward from the pupil at a certain proportion, and this corneal region can be used as the image region covering the corneal region.

[0051] Optionally, the U-Net network localization method can be to input the first image into the U-Net network for pupil and corneal region localization processing to obtain an image region covering the corneal region.

[0052] Optionally, since the anterior surface of the cornea is equivalent to a convex mirror with a radius of curvature of approximately 7.8 mm, ambient light (including natural light, artificial light, and screen light) will form a virtual image on the corneal surface, namely the first Purkinje image (P1). Therefore, the image region covering the corneal area can be segmented from the first image using the PCCR algorithm to obtain the corneal region image. The corneal region image can then be filtered, for example, filtering out the fixed light spots generated by the near-infrared light source module (such as infrared LEDs) while retaining the reflective layer signal carrying environmental information, resulting in a filtered corneal region image, which is then used as the second image.

[0053] In this embodiment, by segmenting the corneal region in the first image and filtering out the light spot information, a second image is obtained, and then subsequent myopia prevention and control processing is performed. This avoids the phenomenon that the image is too large, which would consume too many resources in the myopia prevention and control processing.

[0054] Step S20: Perform spherical distortion correction processing on the second image to obtain the target corneal reflection image; Optionally, the second image can be approximated as the first Pkinje image (hereinafter referred to as the P1 image). Since the original P1 image is an extremely distorted fisheye image, it needs to undergo optical restoration processing for effective scene recognition. This involves performing spherical distortion correction on the second image to obtain the optically restored second image, which is then used as the target corneal reflection image. Alternatively, since the second image carries the reflective layer signal containing environmental information, the target corneal reflective image can be approximated as an environmental mirror image.

[0055] Optionally, step S20, which involves performing spherical distortion correction on the second image to obtain the target corneal reflection image, includes steps b10-b40.

[0056] Step b10: Construct an optical model of the eyeball corresponding to the user wearing the wearable device, and determine the spherical coordinate system corresponding to the optical model of the eyeball; Step b20: Determine the mapping relationship between the spherical coordinate system and the Cartesian coordinate system corresponding to the second image; Step b30: Based on the mapping relationship, each pixel in the second image is mapped into a ray emanating from the shooting end, and a corneal reflection image is constructed based on the intersection of each ray emanating from the eye's optical model. Step b40: Perform detail enhancement processing on the corneal reflection image to obtain the target corneal reflection image.

[0057] Optionally, when performing spherical distortion correction on the second image, three steps can be performed: establishing an eye optical model, inverse mapping transformation, and super-resolution reconstruction.

[0058] Optionally, the eye optical model can be a three-dimensional model of the eyeball representing the user's eye, and the corneal curvature radius R corresponding to different users' eye optical models is different, for example, R is 7.8mm.

[0059] Optionally, eye optical models for different age groups and users can be pre-built and stored in a preset knowledge base, with each eye optical model in the knowledge base corresponding to a different corneal curvature radius.

[0060] Optionally, when establishing an eye optical model, the corneal curvature radius of the user wearing the wearable device can be determined first, and the corresponding eye optical model can be searched in the knowledge base based on the user's corneal curvature radius, and used as the eye optical model corresponding to the user wearing the wearable device.

[0061] Optionally, during the inverse mapping transformation, it is necessary to first determine the spherical coordinate system corresponding to the eye's optical model (e.g., a coordinate system constructed with a point in the eye's optical model as the origin) and the Cartesian coordinate system corresponding to the second image (e.g., a coordinate system constructed with the vertices of the second image as the origin). Then, coordinate alignment processing is performed on the spherical coordinate system and the Cartesian coordinate system to determine the mapping relationship between the spherical coordinate system and the Cartesian coordinate system corresponding to the second image. For example, the mapping relationship can be determined through the world coordinate system. That is, first determine the mapping relationship between the spherical coordinate system and the world coordinate system, then determine the mapping relationship between the Cartesian coordinate system and the world coordinate system, and finally obtain the mapping relationship between the spherical coordinate system and the Cartesian coordinate system.

[0062] Optionally, taking a camera as the capturing end as an example, the coordinates of each pixel in the second image can be converted into the camera ray direction. Normalized camera coordinates can be calculated by combining the camera intrinsic parameter matrix with the pixel coordinates in the second image to obtain the unit direction vector in the camera coordinate system. Then, based on the correspondence between the camera coordinate system (with the camera optical center as the origin) and the world coordinate system (with the corneal center as the origin), the unit direction vector is transformed into the world coordinate system to obtain the camera direction vector. For example, the camera rotation matrix and translation vector can be used, combined with the camera's position relative to the corneal center, to calculate the camera direction vector in the world coordinate system. Then, the reflection point on the corneal surface is calculated using the camera direction vector (i.e., determining the position point where the ray emitted from the camera passes through the corneal surface and using it as the reflection point). Then, the direction of the incident ray is calculated backward based on the reflection point, and the scene is reconstructed based on the direction of the incident ray to obtain the corneal reflection image.

[0063] Optionally, a back-projection transformation algorithm (such as the pupil-center corneal reflection method) can be used to perform a back-projection transformation on the second image to cancel the barrel distortion caused by convex reflection and obtain a corneal reflection image.

[0064] Alternatively, the second image can be directly input into the AI ​​large model, and the AI ​​large model can be used to perform back-projection transformation on the second image to obtain the corneal reflection image.

[0065] Optionally, since corneal reflection images occupy fewer pixels on the original sensor, a GAN (Generative Adversarial Network)-based super-resolution algorithm can be used to enhance the details of the corneal reflection image (i.e., the low-resolution environmental image after distortion correction), restore the object outline and text edges, and obtain the target corneal reflection image.

[0066] In this embodiment, an eye optical model is constructed, and the mapping relationship between the spherical coordinate system corresponding to the eye optical model and the planar rectangular coordinate system corresponding to the second image is determined. Then, the intersection point of the ray and the eye optical model is determined according to the mapping relationship, thereby constructing a corneal reflection image. Then, detail enhancement processing is performed to obtain the target corneal reflection image, thus ensuring the accuracy and effectiveness of the obtained target corneal reflection image.

[0067] Step S30: Perform multi-dimensional dynamic analysis on the target corneal reflection image to obtain multi-dimensional evaluation information; It should be noted that multi-dimensional dynamic analysis includes target object recognition, distance estimation, and illumination detection; Optionally, after obtaining the target corneal reflection image, multi-dimensional eye behavior analysis can be performed. For example, the target corneal reflection image can be used for target object recognition, distance estimation, and illumination detection to achieve multi-dimensional eye behavior analysis of the user.

[0068] Optionally, target object recognition can be performed on the target corneal reflection image to determine whether a target object (such as a mobile phone, tablet, electronic screen, book, etc.) exists in the target corneal reflection image, and the recognition result of the target object recognition can be used as an evaluation information of one dimension.

[0069] Optionally, distance estimation can be performed on the target corneal reflection image. When it is determined that a target object exists in the target corneal reflection image, the distance between the target object and the user's cornea can be estimated, and the estimated distance can be used as an evaluation information in one dimension.

[0070] Optionally, illumination detection can be performed on the target corneal reflection image to determine the corresponding illumination intensity and use it as an evaluation dimensional information.

[0071] Optionally, other dimensions of evaluation and analysis can be performed on the target corneal reflection image. For example, light direction analysis can be performed on the target corneal reflection image to determine the direction of light rays illuminating the target object, thereby identifying the light source corresponding to each light direction and using it as one dimension of evaluation information. For example, user viewing posture analysis can be performed on the target corneal reflection image to determine the user's viewing posture (including viewing angle and viewing position) when viewing the target object.

[0072] Optionally, evaluation information from various dimensions can be aggregated to obtain multi-dimensional evaluation information.

[0073] Optionally, the multi-dimensional assessment information may also include other environmental information corresponding to the target corneal reflection image, such as temperature, humidity, and ambient noise.

[0074] Step S40: In response to the multi-dimensional assessment information not meeting the preset myopia prevention and control conditions, a myopia risk is determined, and a preset prompt message is output to the user indicating the myopia risk.

[0075] It should be noted that the preset myopia prevention and control conditions include at least one of the following: fixation time conditions, fixation distance conditions, and light intensity conditions.

[0076] Optionally, after obtaining multi-dimensional evaluation information relative to the target corneal reflection image, the evaluation information of each dimension in the multi-dimensional evaluation information can be compared with its corresponding evaluation threshold to determine whether the user's current viewing habits pose a risk of myopia. When it is detected that at least one dimension of the multi-dimensional evaluation information does not meet the corresponding condition requirements (such as at least one of the fixation time condition, fixation distance condition, and light intensity condition), it can be considered that the multi-dimensional evaluation information does not meet the preset myopia prevention and control conditions. At this time, it can be considered that the user has a risk of myopia, and a preset prompt message reminding the user of the risk of myopia can be output. The output method can be voice playback, vibration reminder, etc.

[0077] Optionally, after step S30, which involves performing multi-dimensional dynamic analysis on the target corneal reflection image to obtain multi-dimensional evaluation information, at least one of steps c10-c30 is also included.

[0078] Step c10: In response to the multi-dimensional evaluation information including the user is looking at the target object, if the continuous gaze time of the user looking at the target object is greater than the preset gaze time, then it is determined that the multi-dimensional evaluation information does not meet the gaze time condition. Optionally, myopia control can be performed continuously for a certain period of time using wearable devices. Within a continuous period of time (e.g., 5 minutes, 10 minutes, 30 minutes, etc.), multi-dimensional evaluation information, including whether the user is looking at the target object, can be used to determine the time the user is looking at the target object. For example, by performing target object recognition on the target corneal reflection images at different time steps, all target corneal reflection images representing the user looking at the target object can be identified, and it can be determined whether all target corneal reflection images have temporal continuity. If so, the continuous time is taken as the continuous viewing time of the user looking at the target object.

[0079] Optionally, a preset fixation time, such as 20 minutes, can be set in advance. If the user's continuous fixation time on the target object exceeds the preset fixation time, such as the user's continuous fixation time on an electronic screen exceeding 20 minutes, it can be determined that the multi-dimensional assessment information does not meet the fixation time condition, and it can be considered that there is a screen risk. This screen risk may lead to the user's myopia risk, and a preset prompt message will be output to remind the user of the myopia risk.

[0080] Step c20: In response to the multi-dimensional evaluation information including the first distance and / or the second distance, if the first distance and / or the second distance are less than a preset distance threshold, it is determined that the multi-dimensional evaluation information does not meet the gaze distance condition. Optionally, myopia prevention and control can be performed continuously for a certain period of time through wearable devices, and distance estimation can be performed to determine the distance between the user and the target object. For example, when the distance between the user and the electronic screen or book is the first distance and / or the second distance, it can be detected whether the first distance and / or the second distance is less than a preset distance threshold (the distance threshold set by the user in advance, such as 33cm). If it is less than, for example, the user's reading distance (i.e., the first distance or the second distance) is less than 33cm, it can be determined that the multi-dimensional assessment information does not meet the fixation distance condition, and it can be considered that there is a risk of myopia. The user may have a risk of myopia due to the risk of myopia, and a preset prompt message will be output to remind the user of the risk of myopia.

[0081] Step c30: In response to the multi-dimensional evaluation information including light intensity, if the light intensity is less than the preset light intensity threshold, it is determined that the multi-dimensional evaluation information does not meet the light intensity condition.

[0082] Optionally, myopia prevention and control can be performed continuously for a certain period of time through wearable devices, and light detection can be performed to determine the light intensity of the user's environment. The light intensity is compared with a preset light intensity threshold (a light intensity threshold set by the user in advance, such as 300 lux). If the light intensity is less than the preset light intensity threshold, for example, in a user's reading scenario, the light intensity is less than 300 lux, then it can be determined that the multi-dimensional assessment information does not meet the light intensity condition, and it can be considered that there is a light risk. This light risk may lead to myopia risk for the user, and a preset prompt message will be output to remind the user of the myopia risk.

[0083] Optionally, by judging at least one of the following factors—the continuous fixation time of the user looking at the target object, the first distance and / or the second distance between the user and the target object, and the light intensity—it can be determined whether the corresponding fixation time condition, fixation distance condition, and light intensity condition are met. This allows for a comprehensive assessment of the target corneal reflection image from multiple dimensions, thereby improving the accuracy of determining whether the user has a risk of myopia. Upon determination, corresponding prompt information is output, thus enabling precise myopia prevention and control.

[0084] In this embodiment, myopia prevention and control is achieved by utilizing the existing camera on the wearable device. A first image representing the user's eye information is captured, and a second image representing the user's corneal region is extracted from it. Spherical distortion correction processing is then performed to obtain a target corneal reflection image. This target corneal reflection image is then subjected to multi-dimensional dynamic analysis (such as target object recognition, distance estimation, and illumination detection) to obtain multi-dimensional evaluation information. If the multi-dimensional evaluation information does not meet preset myopia prevention and control conditions (such as fixation time, fixation distance, and illumination conditions at least one), a myopia risk is identified, and a preset warning message indicating the myopia risk is output to the user. This allows for accurate prediction using the target corneal reflection image without increasing additional hardware costs. A comprehensive evaluation is performed from multiple dimensions (such as fixation time, fixation distance, and illumination conditions) to improve the accuracy of myopia prevention and control for the user. Therefore, accurate myopia prevention and control can be achieved without increasing the hardware cost of the wearable device.

[0085] Based on the first embodiment of this application, a second embodiment of this application is proposed. In this second embodiment, content that is the same as or similar to the above embodiment can be referred to the above description, and will not be repeated hereafter. Based on this, refer to... Figure 3 In step S30, the step of performing multi-dimensional dynamic analysis on the target corneal reflection image to obtain multi-dimensional evaluation information also includes steps d10-d30.

[0086] Step d10, in response to multi-dimensional dynamic analysis including target object recognition, high-frequency feature recognition is performed on the target corneal reflection image, and the existence of a first image region representing a rectangular outline in the target corneal reflection image is detected. Step d20: In response to the presence of high-frequency features in the target corneal reflection image and the existence of a first image region, a second image region in the target corneal reflection image other than the first image region is determined, and it is detected whether the image parameters of the first image region and the image parameters of the second image region match, and whether the color temperature information of the ROI region in the target corneal reflection image is less than a preset color temperature threshold. In step d30, in response to the mismatch between the image parameters of the first image region and the image parameters of the second image region, and the color temperature information of the ROI region being less than the preset color temperature threshold, it is determined that the multi-dimensional evaluation information includes the target object being viewed by the user.

[0087] Optionally, when identifying target objects from the corneal reflection image, high-frequency feature recognition can be performed. For example, if a user stares at the sky or a wall for an extended period, the sky or wall can be assumed to be stationary, therefore the corneal reflection image will lack high-frequency features. However, if a user stares at an electronic screen (such as a mobile phone) for an extended period, such as watching a video on their phone, the pixel features in the corneal reflection image corresponding to the mobile phone screen can be considered high-frequency features because the image on the screen is constantly updating.

[0088] Optionally, when performing high-frequency feature recognition on the target corneal reflection image, the recognition process can be performed by combining images from other video frames adjacent to the target corneal reflection image. For example, if the features of a pixel location or pixel region are continuously updated and changed in the target corneal reflection image of 10 consecutive frames, the features of that pixel location or pixel region can be used as high-frequency features.

[0089] Alternatively, since mobile phones, tablets, or displays have high brightness and regular geometry, the Canny edge detection operator combined with the Hough Transform can be used to find closed quadrilateral contours (such as rectangular contours) in the target corneal reflection image.

[0090] Optionally, the Region of Interest (ROI) in the target corneal reflection image can be determined. For example, a first image representing the user's eye information can be determined first, and the ROI region in the first image (hereinafter referred to as the first ROI region) can be determined according to the relevant instructions input by the user. Then, the ROI region corresponding to the first ROI region can be found in the target corneal reflection image and used as the ROI region in the target corneal reflection image.

[0091] Alternatively, the first image region that represents the rectangular outline can be directly used as the ROI region.

[0092] Optionally, the target object can be an object that the user is looking at, an electronic screen of an electronic device (such as a mobile phone screen), or other objects such as a book.

[0093] Optionally, if high-frequency features are detected in the target corneal reflection image, it can be assumed that the object the user is viewing is changing rapidly. If a first image region representing a rectangular outline is detected in the target corneal reflection image, image parameters (such as brightness, grayscale value, chromaticity, etc., with brightness as an example below) of this first image region and a second image region in the target corneal reflection image can be obtained to determine whether the user may be looking at a target object, such as an electronic screen. Furthermore, since electronic screens typically contain high-energy blue light, their spectral characteristics are manifested as specific grayscale ratios on the image sensor. Therefore, the grayscale ratio of the ROI region in the target corneal reflection image can be extracted and used as a chromaticity feature.

[0094] Optionally, the color temperature features of the ROI region in the target corneal reflection image can be extracted and used as the color temperature information of the ROI region. Then, it can be detected whether the color temperature information is less than a preset color temperature threshold (a pre-set color temperature threshold, such as grayscale value).

[0095] Optionally, if the image parameters of the first image region and the second image region do not match, and the color temperature information of the ROI region is less than a preset color temperature threshold (i.e., the color temperature can be considered to be cool), then it can be determined that the user is looking at the target object, such as an electronic screen. For example, if a rectangular outline is detected in the target corneal reflection image, and its brightness is significantly higher than the background brightness, and the color temperature is cool, then it can be determined that the user wearing the wearable device is looking at an electronic screen.

[0096] In this embodiment, when identifying a target object in a corneal reflection image, if it is determined that there are high-frequency features in the corneal reflection image, a first image region representing a rectangular outline exists, the image parameters of the first image region and the image parameters of the second image region do not match, and the color temperature information of the ROI region is less than a preset color temperature threshold, it is determined that the user is looking at the target object, and thus the user's gaze information can be accurately identified.

[0097] Optionally, distance estimation includes coarse distance measurement, and the shooting end includes a left-eye camera and a right-eye camera.

[0098] Optionally, the left eye camera can be a left-eye endoscopic camera built into the wearable device, and the right eye camera can be a right-eye endoscopic camera built into the wearable device.

[0099] Optionally, step S30, which involves performing multi-dimensional dynamic analysis on the target corneal reflection image to obtain multi-dimensional evaluation information, further includes steps e10-e20.

[0100] Step e10, in response to multi-dimensional dynamic analysis including coarse distance measurement, determines the left-eye image and right-eye image corresponding to the target corneal reflection image, wherein the left-eye image and right-eye image are images captured by the left-eye camera and the right-eye camera respectively at the same time step; Step e20: Calculate the binocular eye gaze angle based on the left-eye and right-eye images, determine the first distance between the user and the target object based on the binocular eye gaze angle, and determine multi-dimensional evaluation information including the first distance, wherein the larger the binocular eye gaze angle, the smaller the first distance.

[0101] Optionally, distance estimation can be performed on the target corneal reflection image to determine the distance between the user and the target object, such as the distance between the user and an electronic screen.

[0102] Optionally, when estimating the distance from the target corneal reflection image, a binocular convergence method can be used for coarse distance measurement.

[0103] Optionally, a first image representing the user's eye information can be acquired by the left-eye camera and the right-eye camera at the same time step, and the first image captured by the left-eye camera at the same time step as the target corneal reflection image can be selected as the left-eye image, and the first image captured by the right-eye camera at the same time step as the target corneal reflection image can be selected as the right-eye image. Then, the angle between the user's eyes can be calculated using the left-eye image and the right-eye image.

[0104] The vectors between the pupil center and the corneal reflection point are calculated by using the left and right eye cameras respectively to obtain the direction of the gaze of both eyes. Then, the convergence angle is estimated based on the direction of the gaze of both eyes and used as the angle between the gaze of both eyes.

[0105] Optionally, after determining the angle of gaze between the two eyes, the distance between the user and the target object (such as an electronic screen) can be estimated based on the angle of gaze between the two eyes, and this distance can be used as the first distance. The larger the angle of gaze between the two eyes, the smaller the first distance.

[0106] Optionally, a first mapping table representing the mapping relationship between different binocular line-of-sight angles and distances can be pre-set. The corresponding distance can be looked up in the first mapping table based on the binocular line-of-sight angle corresponding to the target corneal reflection image, and used as the first distance. Alternatively, the first distance can be estimated by combining the binocular line-of-sight angles with the principle of triangulation.

[0107] In this embodiment, when performing a coarse distance measurement on the target corneal reflection image, the angle between the two eyes' lines of sight can be calculated based on the left and right eye images to determine the first distance between the user and the target object, thereby ensuring the accuracy and effectiveness of the determined first distance.

[0108] Optionally, distance estimation includes precise distance measurement.

[0109] Optionally, step S30, which involves performing multi-dimensional dynamic analysis on the target corneal reflection image to obtain multi-dimensional evaluation information, further includes steps f10-f20.

[0110] Step f10, in response to multi-dimensional dynamic analysis including distance precision measurement, identifies a third image region representing the target object in the target corneal reflection image and determines the pixel size parameter information of the third image region; Step f20: Determine the actual size parameter information of the target object. Based on the comparison result between the pixel size parameter information and the actual size parameter information, determine the second distance between the user and the target object, and determine the multi-dimensional evaluation information including the second distance.

[0111] Optionally, distance estimation can be performed on the target corneal reflection image to determine the distance between the user and the target object, such as the distance between the user and an electronic screen.

[0112] Optionally, when estimating the distance of a target corneal reflection image, the mirror ratio method can be used for precise distance measurement.

[0113] Optionally, when it is determined that a target object exists in the corneal reflection image, a third image region representing the target object can be determined (for example, a first image region representing a rectangular outline). Then, the pixel size parameter information of the third image region (such as pixel area, pixel perimeter, etc.) is determined, for example, the pixel area ratio of the target object (such as a mobile phone) in the corneal reflection image.

[0114] Optionally, visual size parameters of the target object can be determined, such as the standard size of a mobile phone, including its perimeter and area.

[0115] Optionally, based on the formula of the convex mirror imaging principle ( Where u is the object distance, v is the image distance, f is the lens focal length, and f = R / 2, where R is the corneal curvature radius (e.g., 7.8 mm). It can be known that the image size of an object is inversely proportional to the object distance. Therefore, the second distance between the user and the target object can be determined based on the comparison between the pixel size parameter information and the actual size parameter information.

[0116] Optionally, a second mapping table can be set to represent the mapping relationship between the comparison results of pixel size parameter information and actual size parameter information and the distance, and the corresponding distance can be queried in the second mapping table and used as the second distance.

[0117] In this embodiment, when performing distance measurement on the target corneal reflection image, the second distance between the user and the target object can be determined based on the comparison between the pixel size parameter information corresponding to the target object and the actual size parameter information of the target object. This allows for monocular assisted ranging by utilizing the size change of the target corneal reflection image (such as the size change of the third image region in the target corneal reflection image), ensuring the accuracy and effectiveness of the determined second distance.

[0118] Optionally, in step S30, the step of performing multi-dimensional dynamic analysis on the target corneal reflection image to obtain multi-dimensional evaluation information further includes step g10.

[0119] Step g10, in response to multi-dimensional dynamic analysis including illumination detection, determines the first shooting time of the first image corresponding to the target corneal reflection image, acquires the illumination intensity of the environment where the wearable device is located at the first shooting time, and determines multi-dimensional evaluation information including illumination intensity.

[0120] Optionally, the illuminance value (Lux) of the environment in which the user wearing the wearable device is located can be calculated using data collected by the photoresistor in the wearable device, and used as the light intensity of the environment in which the wearable device is located.

[0121] Optionally, color and brightness analysis can be performed on the target corneal reflection image, and the light intensity can be determined based on the different color and brightness values ​​obtained from the analysis. This allows for the accurate identification of the light intensity information of the user's environment, such as accurately identifying risky scenarios like "reading in a dim environment" or "direct sunlight".

[0122] Optionally, the capture time of the first image corresponding to the target corneal reflection image can be determined (i.e., the first capture time, such as 3 pm), and the illuminance value collected at that capture time can be determined based on the photoresistor and used as the light intensity.

[0123] In this embodiment, the first shooting time is determined based on the target corneal reflection image, and the light intensity collected at the first shooting time is determined, thereby ensuring the accuracy and effectiveness of the obtained light intensity. This enables myopia prevention and control assessment from the perspective of light intensity, determining whether the user has a risk of myopia, and improving the accuracy of myopia prevention and control.

[0124] In addition, to aid in understanding the myopia prevention principle based on wearable devices in this embodiment, examples are provided below.

[0125] For example, such as Figure 4 As shown, when wearable devices begin myopia prevention, image acquisition (obtaining eye images) can be performed. For example, a first image representing the user's eye information (i.e., the eye image) can be acquired using a left-eye intraocular camera and / or a right-eye intraocular camera. Then, ROI extraction is performed to locate the corneal region, i.e., determining the ROI region in the first image and locating the corneal region within it. Next, image preprocessing is performed on the first image (filtering out light spots + extracting the P1 image), i.e., extracting a second image representing the user's corneal region from the first image. For example, determining the image region covering the corneal region in the first image and segmenting this region to obtain the corneal region image. Light spot information in the corneal region image is then filtered out to obtain the second image. Spherical distortion correction (back-projection transformation and super-resolution reconstruction) is then performed on the second image to obtain the target corneal reflection image. A parallel analysis module is then used to perform multi-dimensional dynamic analysis on the target corneal reflection image, such as screen recognition (edge ​​detection + color temperature analysis), to determine whether the user is looking at an electronic screen. For example, distance estimation (convergence angle + mirror ratio calculation) involves coarse and fine distance measurements to determine the distance between the user and the electronic screen (e.g., a first or second distance). Light monitoring (photoresistance statistics) determines light intensity, thereby identifying the lighting information of the user's environment. A comprehensive risk assessment is then performed to determine if thresholds are exceeded. This involves evaluating whether multi-dimensional assessment information meets preset myopia prevention conditions (e.g., at least one of fixation time, fixation distance, and light intensity). If so, the user is identified as having a myopia risk, triggering feedback (vibration alert / app push notification) to alert the user. If not, data is recorded and the model is updated (i.e., a second myopia prevention process is performed) until the next frame is analyzed, continuously implementing myopia prevention based on wearable devices.

[0126] Furthermore, this application provides a wearable device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, which are executed by the at least one processor to enable the at least one processor to perform the myopia prevention and control method based on the wearable device in the first embodiment described above.

[0127] The following is for reference. Figure 5The figure illustrates a structural schematic diagram suitable for implementing the embodiments of this application. The wearable devices in the embodiments of this application may include, but are not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. The devices shown in the figure are merely examples and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0128] Wearable devices may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.) that can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for device operation. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1003 including, for example, magnetic tape, hard disk, etc.; and communication devices 1009. Communication device 1009 allows the wearable device to communicate wirelessly or wiredly with other devices to exchange data. While wearable devices with various systems are shown in the figures, it should be understood that implementation or possession of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0129] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0130] The wearable device provided in this application, employing the myopia prevention and control method based on wearable devices described in the above embodiments, can solve the technical problem of how to achieve accurate myopia prevention and control without increasing the hardware cost of the wearable device. Compared with the prior art, the beneficial effects of the wearable device provided in this application are the same as those of the myopia prevention and control method based on wearable devices provided in the above embodiments, and other technical features of the wearable device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0131] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0132] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0133] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the myopia prevention and control method based on a wearable device in the above embodiments.

[0134] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0135] The aforementioned computer-readable storage medium may be included in the wearable device; or it may exist independently and not assembled into the wearable device.

[0136] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by a wearable device, enable the wearable device to perform the steps and processes in the aforementioned myopia prevention and control method based on the wearable device.

[0137] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0138] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0139] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0140] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the aforementioned myopia prevention and control method based on wearable devices. This solves the technical problem of achieving accurate myopia prevention and control without increasing the hardware cost of the wearable device. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the myopia prevention and control method based on wearable devices provided in the above embodiments, and will not be repeated here.

[0141] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the wearable device-based myopia prevention and control method described above.

[0142] The computer program product provided in this application solves the technical problem of how to achieve accurate myopia prevention and control without increasing the hardware cost of wearable devices. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the myopia prevention and control method based on wearable devices provided in the above embodiments, and will not be repeated here.

[0143] The above are only some embodiments of this application and do not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for myopia prevention and control based on wearable devices, characterized in that, The myopia prevention method based on wearable devices is applied to wearable devices, which include at least one camera, and includes the following steps: Acquire a first image representing the user's eye information, and extract a second image representing the user's corneal region from the first image; The second image is subjected to spherical distortion correction processing to obtain the target corneal reflection image; The target corneal reflection image is subjected to multi-dimensional dynamic analysis to obtain multi-dimensional evaluation information, wherein the multi-dimensional dynamic analysis includes target object recognition, distance estimation and illumination detection; In response to the multi-dimensional assessment information not meeting the preset myopia prevention and control conditions, a myopia risk is determined, and a preset prompt message is output to the user indicating the myopia risk. The preset myopia prevention and control conditions include at least one of fixation time conditions, fixation distance conditions, and light intensity conditions.

2. The myopia prevention and control method based on wearable devices as described in claim 1, characterized in that, The step of performing spherical distortion correction processing on the second image to obtain the target corneal reflection image includes: Construct an optical model of the eyeball corresponding to the user wearing the wearable device, and determine the spherical coordinate system corresponding to the optical model of the eyeball; Determine the mapping relationship between the spherical coordinate system and the Cartesian coordinate system corresponding to the second image; Based on the mapping relationship, each pixel in the second image is mapped into a ray emanating from the shooting end, and a corneal reflection image is constructed based on the intersection of each ray emanating from the eye optical model; The corneal reflection image is enhanced with detail to obtain the target corneal reflection image.

3. The myopia prevention and control method based on wearable devices as described in claim 1, characterized in that, The step of performing multi-dimensional dynamic analysis on the target corneal reflection image to obtain multi-dimensional evaluation information includes: In response to the multi-dimensional dynamic analysis including target object recognition, high-frequency feature recognition is performed on the target corneal reflection image, and the existence of a first image region representing a rectangular outline in the target corneal reflection image is detected. In response to the presence of high-frequency features in the target corneal reflection image and the existence of the first image region, a second image region other than the first image region in the target corneal reflection image is determined, and it is detected whether the image parameters of the first image region and the image parameters of the second image region match, and whether the color temperature information of the ROI region in the target corneal reflection image is less than a preset color temperature threshold. If the image parameters of the first image region and the image parameters of the second image region do not match, and the color temperature information of the ROI region is less than a preset color temperature threshold, then it is determined that the multi-dimensional evaluation information includes the target object being viewed by the user.

4. The myopia prevention and control method based on wearable devices as described in claim 1, characterized in that, The distance estimation includes a coarse distance measurement, and the shooting device includes a left-eye camera and a right-eye camera. The step of performing multi-dimensional dynamic analysis on the target corneal reflection image to obtain multi-dimensional evaluation information includes: In response to the multi-dimensional dynamic analysis including coarse distance measurement, the left-eye image and the right-eye image corresponding to the target corneal reflection image are determined, wherein the left-eye image and the right-eye image are images captured by the left-eye camera and the right-eye camera respectively at the same time step; The binocular eye gaze angle is calculated based on the left-eye image and the right-eye image. A first distance between the user and the target object is determined based on the binocular eye gaze angle. The multi-dimensional evaluation information includes the first distance. The larger the binocular eye gaze angle, the smaller the first distance.

5. The myopia prevention and control method based on wearable devices as described in claim 1, characterized in that, The distance estimation includes precise distance measurement. The step of performing multi-dimensional dynamic analysis on the target corneal reflection image to obtain multi-dimensional evaluation information includes: In response to the multi-dimensional dynamic analysis including distance precision measurement, a third image region representing the target object in the target corneal reflection image is identified, and the pixel size parameter information of the third image region is determined; The actual size parameter information of the target object is determined. Based on the comparison result between the pixel size parameter information and the actual size parameter information, a second distance between the user and the target object is determined, and the multi-dimensional evaluation information includes the second distance.

6. The myopia prevention and control method based on wearable devices as described in claim 1, characterized in that, The step of performing multi-dimensional dynamic analysis on the target corneal reflection image to obtain multi-dimensional evaluation information further includes: In response to the multi-dimensional dynamic analysis including illumination detection, the first shooting time of the first image corresponding to the target corneal reflection image is determined, the illumination intensity of the environment where the wearable device is located is acquired at the first shooting time, and the multi-dimensional evaluation information includes the illumination intensity.

7. The myopia prevention and control method based on wearable devices as described in any one of claims 1-6, characterized in that, After the step of performing multi-dimensional dynamic analysis on the target corneal reflection image to obtain multi-dimensional evaluation information, at least one of the following is included: In response to the multi-dimensional evaluation information including that the user is looking at the target object, if the continuous gaze time of the user looking at the target object is greater than the preset gaze time, then it is determined that the multi-dimensional evaluation information does not meet the gaze time condition. In response to the multi-dimensional evaluation information including a first distance and / or a second distance, if the first distance and / or the second distance is less than a preset distance threshold, it is determined that the multi-dimensional evaluation information does not meet the gaze distance condition; In response to the multi-dimensional evaluation information including light intensity, if the light intensity is less than a preset light intensity threshold, it is determined that the multi-dimensional evaluation information does not meet the light intensity condition.

8. The myopia prevention and control method based on wearable devices as described in any one of claims 1-6, characterized in that, The step of extracting a second image representing the user's corneal region from the first image includes: The image region covering the corneal area in the first image is determined, and the image region covering the corneal area is segmented to obtain a corneal area image. The light spot information in the corneal area image is filtered out to obtain a second image.

9. A wearable device, characterized in that, The wearable device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the myopia prevention and control method based on the wearable device as described in any one of claims 1 to 8.

10. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the myopia prevention and control method based on a wearable device as described in any one of claims 1 to 8.

Images