Wearable head eye tracking device, method, processing device and system

Abstract

The embodiment of the application provides a wearable head eye movement tracking device, method, processing device and system, and belongs to the technical field of life science. The wearable head eye movement tracking device comprises a device outer frame, a hole is arranged in the middle of the device outer frame, the hole is through from top to bottom, and ear supports protruding downward are arranged on the two sides of the device outer frame, so that the device outer frame is fixed on the head of a dog; and the wearable head eye movement tracking device further comprises at least one eyeball tracking camera, the eyeball tracking camera is used for acquiring an eye image containing the eyes of the dog, and the eye image is used for being sent to a processing device in communication connection with the wearable head eye movement tracking device, so that the processing device identifies the eye image and obtains a target pupil diameter of the eyes of the dog. The application can realize tracking of eye information of a dog with a larger body size and stronger free activity without space limitation, and improves the accuracy of the collected eye information of the dog.

Description

Technical Field

[0001] This application relates to the field of life science technology, and in particular to a wearable head eye-tracking device, method, processing device and system. Background Technology

[0002] Dogs' eyes typically reveal a wealth of information; for example, a sudden dilation of the pupils can be caused by fear. Therefore, monitoring changes in eye information (i.e., eye tracking) is of great importance, and eye trackers are commonly used to track canine eye movements.

[0003] However, most dogs are large and highly mobile. The eye-tracking devices used in related technologies can only detect dogs in a fixed position or can only detect dogs by deploying fixed cameras in a preset space. Either way, they limit the range of dogs' free movement, which greatly restricts the dogs' spatial behavior and makes it impossible to know their eye information in various scenarios, thus affecting the accuracy of the collected dog eye information. Summary of the Invention

[0004] The main objective of this application is to propose a wearable head eye-tracking device, method, processing device, and system that can track the eye information of large, highly mobile dogs under unrestricted conditions, thereby improving the accuracy of the collected canine eye information.

[0005] To achieve the above objectives, a first aspect of this application provides a wearable head eye-tracking device, comprising:

[0006] The device frame has an opening in the middle that runs vertically through the frame. On both sides of the device frame are downwardly protruding ear supports that are used to mount the dog's ears to fix the device frame to the dog's head. The front of the device frame also has a protruding camera support.

[0007] At least one eye-tracking camera is mounted on the camera mount and is used to photograph dogs wearing the wearable head-eye tracking device to obtain eye images including the dog's eyes.

[0008] The eye image is sent to a processing device that is communicatively connected to the wearable head eye-tracking device, so that the processing device can identify the pupil of the canine eye in the eye image and extract the target pupil diameter of the canine eye.

[0009] In some embodiments, the device frame is provided with an adjustable headband that wraps around the opening and has a fixing component for fixing the adjustable size of the headband.

[0010] In some embodiments, the camera bracket is a small-volume bracket, and the eye-tracking camera is disposed at one end of the camera bracket away from the outer frame of the device, and the eye-tracking camera is provided with an infrared device.

[0011] In some embodiments, the device frame is further provided with an elastic contact surface that surrounds the headband so that the device frame contacts the dog when it is fixed to the dog's head.

[0012] In some embodiments, the device frame is further provided with a panoramic camera, which is aligned with the dog's eye orientation to record the dog's movement trajectory and the external environment information corresponding to the eye image.

[0013] To achieve the above objectives, a second aspect of this application proposes a wearable head eye-tracking method, applied in a processing device. The processing device is communicatively connected to a wearable head eye-tracking device. The wearable head eye-tracking device includes a device frame and at least one eye-tracking camera. An opening is provided in the middle of the device frame, and the opening is vertically continuous. Ear supports protruding downwards are provided on both sides of the device frame. The ear supports are used to mount on the ears of a dog to fix the device frame to the dog's head. A camera support protruding forward is also provided on the front side of the device frame. The eye-tracking camera is mounted on the camera support and is used to photograph a dog wearing the wearable head eye-tracking device to obtain an eye image including the dog's eyes.

[0014] The method includes:

[0015] Acquire the eye images sent by the wearable head eye-tracking device;

[0016] The pupils of canine eyes in the eye images are identified, and the target pupil diameter of the canine eyes is extracted.

[0017] In some embodiments, identifying the pupil of a canine eye in the eye image and extracting the target pupil diameter of the canine eye includes:

[0018] The eye image is input into a pre-trained pupil recognition model to extract the first diameter information of the canine eye in the eye image;

[0019] The eye image is segmented according to a preset segmentation threshold to extract the second diameter information of the canine eye in the eye image;

[0020] The target pupil diameter of a canine eye is obtained by fusing the first diameter information and the second diameter information.

[0021] In some embodiments, inputting the eye image into a pre-trained pupil recognition model to extract the first diameter information of the canine eye in the eye image includes:

[0022] The eye image is input into a pre-trained deep learning model to obtain the pupil center point of the eye image;

[0023] Using the center point of the pupil as a reference, select any one of the outer ring feature points of the pupil in the pupil area, and determine other outer ring feature points of the pupil according to a preset interval angle. Repeat this operation until the number of outer ring feature points of the pupil reaches a preset number.

[0024] Based on the pupil center point and the pupil outer ring feature points, the eye image is subjected to a first ellipse fitting process to obtain the first diameter information of the canine eye in the eye image.

[0025] In some embodiments, the step of performing image segmentation processing on the eye image according to a preset segmentation threshold to extract the second diameter information of the canine eye in the eye image includes:

[0026] The eye image is segmented into multiple sub-images. If any sub-image is within a preset segmentation threshold, the sub-image is determined to be a set of canine eye images.

[0027] A second ellipse fitting process is performed on the set of canine eye images to obtain the second diameter information of the canine eyes in the eye images.

[0028] In some embodiments, obtaining the target pupil diameter of a canine eye by fusing the first diameter information and the second diameter information includes:

[0029] Calculate the first noise of the first diameter information, and determine the first state vector and the first covariance matrix corresponding to the current time based on the first noise;

[0030] Based on the first diameter information, the first state vector, and the first covariance matrix, the first state vector and the first covariance matrix corresponding to the next time step after the update are obtained.

[0031] Calculate the second noise of the second diameter information, and determine the second state vector and the second covariance matrix corresponding to the current time based on the second noise;

[0032] Based on the second diameter information, the second state vector, and the second covariance matrix, the updated second state vector and the second covariance matrix for the next time step are obtained.

[0033] The target pupil diameter is determined based on the updated first covariance matrix and the second covariance matrix.

[0034] In some embodiments, determining the target pupil diameter based on the updated first covariance matrix and the second covariance matrix includes:

[0035] Based on the first noise and the second noise, the target covariance at the current time is calculated;

[0036] Based on the first state vector and the second state vector, the target state vector corresponding to the current time is calculated, and based on the first covariance matrix and the second covariance matrix, the target covariance matrix corresponding to the current time is calculated.

[0037] Based on the target covariance, the updated target covariance and target covariance matrix are obtained at the next time step;

[0038] The pupil weight value is determined based on the proportion of the target covariance matrix, and the target state vector is adjusted based on the pupil weight value. The target pupil diameter is then determined based on the adjusted target state vector.

[0039] In some embodiments, the wearable head eye tracking device further includes a panoramic camera, wherein the panoramic camera is aligned with the dog's eye orientation to record the dog's movement trajectory and external environmental information corresponding to the eye images;

[0040] The panoramic camera is used to send the collected eye images and external environment information to the processing device, so that the processing device can process the eye images and external environment information to obtain image display information for display.

[0041] To achieve the above objectives, a third aspect of this application provides a processing device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the wearable head eye tracking device described in the first aspect embodiment and the wearable head eye tracking method described in the second aspect embodiment.

[0042] To achieve the above objectives, a fourth aspect of the present application provides an eye-tracking system, which includes the wearable head eye-tracking device described in the first aspect embodiment and the processing device described in the third aspect embodiment.

[0043] This application provides a wearable head eye-tracking device, method, processing device, and system. The wearable head eye-tracking device includes a device frame with a through-hole in the center. Ear supports protruding downwards are located on both sides of the frame, supporting the dog's ears to secure the frame to the dog's head. A protruding camera bracket is also located on the front of the frame. The device further includes at least one eye-tracking camera mounted on the camera bracket. This camera captures images of the dog wearing the device, including its eyes. These images are then sent to a processing device communicatively connected to the device, enabling the processing device to identify the pupils in the images and extract the target pupil diameter. In this way, larger dogs with a wider range of movement can move freely in any scenario after being equipped with this wearable head eye-tracking device. At the same time, the wearable head eye-tracking device can collect eye information of dogs in real time during their activities, thus completing the tracking of dog eye information under conditions without spatial restrictions and improving the accuracy of the collected dog eye information.

[0044] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0045] Figure 1 This is an optional structural diagram of the wearable head eye-tracking device provided in the embodiments of this application;

[0046] Figure 2 This is a schematic diagram of an optional hardware connection for the wearable head eye-tracking device provided in this application embodiment;

[0047] Figure 3 This is an optional flowchart of the wearable head eye-tracking method provided in the embodiments of this application;

[0048] Figure 4 yes Figure 3 A flowchart of an implementation of step S102 in the process;

[0049] Figure 5 yes Figure 4A flowchart of an implementation of step S201;

[0050] Figure 6 yes Figure 4 A flowchart of an implementation of step S202 in the process;

[0051] Figure 7 This is another optional flowchart of the wearable head eye-tracking device provided in the embodiments of this application;

[0052] Figure 8 yes Figure 4 A flowchart of an implementation of step S203;

[0053] Figure 9 yes Figure 8 A flowchart of the implementation of step S505 in the process;

[0054] Figure 10 This is a schematic diagram of the hardware structure of the processing device provided in the embodiments of this application.

[0055] Reference numerals: 10 for device frame, 11 for ear bracket, 12 for camera bracket, 13 for eye-tracking camera, and 14 for panoramic camera. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0057] Furthermore, unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit the scope of this application.

[0058] The study of canine eye information is an exploration of the canine visual system, which is of great significance to fields such as biomedicine, ecology, and behavior. By observing subtle changes in a dog's eyes, one can often obtain a wealth of information about its emotions, arousal state, cognitive abilities, and attention levels. For example, a sudden dilation of a canine pupil may be due to fear. Therefore, monitoring changes in eye information (i.e., eye tracking) is of great importance in fields such as neuroscience, psychology, and psychiatry. Currently, eye trackers are commonly used to track canine eye information.

[0059] However, most dogs are large and highly mobile. The eye-tracking devices used in related technologies can only detect dogs in a fixed position or can only detect dogs by deploying fixed cameras in a preset space. Either way, they limit the range of dogs' free movement, which greatly restricts the dogs' spatial behavior and makes it impossible to know their eye information in various scenarios, thus affecting the accuracy of the collected dog eye information.

[0060] For example, using wired connections to fixed observation devices for monitoring severely restricts the dog's movement. Furthermore, the complex restraints can easily cause panic in the dog, resulting in the collected eye information not being obtained from the dog's natural state, thus reducing the accuracy of the collected canine eye data. Alternatively, when the dog is within a pre-defined space, multiple cameras can be deployed to capture eye information; however, this method keeps the dog in the same scene, lacking detection of other scenes, which also reduces the accuracy of the collected canine eye data.

[0061] Based on this, embodiments of this application provide a wearable head eye tracking device, method, processing device, and system, which can track the eye information of large dogs with strong free movement under unrestricted conditions, thereby improving the accuracy of the collected canine eye information.

[0062] The wearable head eye tracking device, method, processing device and system provided in the embodiments of this application are specifically described through the following embodiments. First, the wearable head eye tracking device in the embodiments of this application is described.

[0063] like Figure 1 As shown, Figure 1 This is an optional structural diagram of a wearable head eye-tracking device provided in the embodiments of this application. The wearable head eye-tracking device includes a device frame 10, with an opening in the middle of the device frame 10 that runs vertically through the opening. Ear supports 11 protruding downwards are respectively provided on both sides of the device frame 10. The ear supports 11 are used to mount on the dog's ears to fix the device frame 10 on the dog's head. A protruding camera support 12 is also provided on the front side of the device frame 10.

[0064] In some embodiments, the main body of the wearable head eye-tracking device (hereinafter referred to as the "tracking device") is the device frame 10, which is continuous vertically, allowing it to be easily placed on a dog's head. In related technologies, to secure the eye-tracking device to a dog's head, a vertical headband is typically added. However, such eye-tracking devices increase the dog's discomfort and hinder the collection of eye images. Therefore, the wearable head eye-tracking device in this embodiment adds ear supports 11 (also called ear fixators) on both sides. Compared to related eye-tracking devices, using the ear supports 11 stabilizes the tracking device on the dog's head while minimizing contact between the device and the dog, increasing the dog's comfort, reducing discomfort, and allowing the dog to be in a more natural state for better eye information collection.

[0065] Understandably, the through-type device frame 10 also allows more of the dog's skin tissue to be exposed, making it easier for operators to perform procedures such as electrode implantation.

[0066] Furthermore, the tracking device can also be equipped with a Global Positioning System (GPS) to record the dog's movement trajectory while wearing the tracking device. This trajectory consists of multiple footprint points on a satellite map. By recording the dog's movement trajectory during free movement through the GPS positioning system, the analysis of the dog's eye information can be aided. For example, when the dog passes through a park or a crowded area, the collected eye information can show that the dog's pupils are significantly dilated and the outer ring of the pupil is more rounded, indicating that the dog will be more excited when passing through these places.

[0067] In some embodiments, the tracking device is further provided with at least one eye-tracking camera 13, which is mounted on the camera bracket 12. The eye-tracking camera 13 is used to take pictures of dogs wearing wearable head eye-tracking devices to obtain eye images including the dog's eyes.

[0068] The eye images are sent to a processing device that is connected to a wearable head-eye tracking device, so that the processing device can identify the pupils of the canine eyes in the eye images and extract the target pupil diameter of the canine eyes.

[0069] In some embodiments, only one eye-tracking camera 13 may be provided. In this case, the eye-tracking camera 13 can capture images of both eyes of the dog, so that both eyes of the dog are present in the same eye image. Or, as... Figure 1As shown, two eye-tracking cameras 13 can be set up. The left eye tracking device is used to photograph the dog's left eye, and the right eye tracking device is used to photograph the dog's right eye. This ensures the focus of the shooting and also makes it easier to process the eye information later.

[0070] Furthermore, the wearable eye-tracking device can be wirelessly connected to a processing device, which can then recognize and process the received eye images. Moreover, the processing device can be placed in other environments, such as an operator's laboratory, thus freeing the dog from the weight of the tracking device, reducing its burden and allowing for greater freedom of movement.

[0071] Furthermore, to ensure wearing comfort and stability, the device frame 10, ear bracket 11, and camera bracket 12 can be made of lightweight materials, such as acrylonitrile-butadiene-styrene copolymer (ABS).

[0072] Furthermore, such as Figure 2 As shown, Figure 2 This is an optional hardware connection diagram of the wearable head eye-tracking device provided in the embodiments of this application. In order to save on wiring, the three video streams of the panoramic camera, the left eye-tracking camera and the right eye-tracking camera can be integrated together, and the three USB video stream sub-signals can be transmitted to the processing device so that the processing device can process the received signals.

[0073] Furthermore, the processing device can be a smartphone, tablet, laptop, desktop computer, or a standalone physical server. It can also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. It should be noted that the processing device can take many other forms, but is not limited to the above forms, and can be specifically configured according to the actual operating conditions.

[0074] Furthermore, since the eye image contains the dog's eyes, the processing device can obtain information related to the dog's eyes, such as pupil position and pupil diameter, after processing the eye image. By analyzing the pupil position and pupil diameter at different times and in different scenarios, the behavior and potential emotions of dogs can be studied.

[0075] Furthermore, the wearable head eye-tracking device provided in this application embodiment adopts a quick-plug design for each part, so as to disassemble and replace related components, thereby improving the convenience of maintenance and upgrade of the tracking device.

[0076] It should be noted that the wearable head eye tracking device provided in this application embodiment can also be applied to other larger animals, such as orangutans, pigs or cows. This application embodiment is only described with reference to preferred embodiments and does not make any specific limitations.

[0077] It should be noted that in all specific embodiments of this application, when processing data related to the wearer's identity or characteristics, such as wearer information, wearer behavior data, wearer historical data, and wearer location information, permission or consent from the corresponding management personnel is obtained first. For example, when collecting eye images of dogs for identification and data processing, permission or consent from their owners is obtained. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. Additionally, when embodiments of this application require access to sensitive personal information of relevant management personnel, the necessary wearer-related data for the proper functioning of these embodiments is obtained only after obtaining the separate permission or consent of the relevant management personnel.

[0078] In some embodiments, the device frame 10 is provided with an adjustable headband that wraps around an opening and has a fixing component for fixing the adjustable size of the headband.

[0079] Furthermore, the headband (not shown in the figure) can be made of a stretchable material to accommodate dogs of different sizes. For example, when equipping a dog with the wearable head eye tracking device, the headband can be used to roughly estimate the size of the dog's head, and the headband can be fixed to a certain adjustable size using the fixing components. In this way, the wearable head eye tracking device can provide a comfortable head circumference size for different dogs, and can also stabilize the wearable head eye tracking device on the heads of dogs of different sizes.

[0080] Furthermore, the headband wraps around the opening, reducing additional space requirements and preventing obstruction of the dog's vision when equipped with the tracking device.

[0081] In some embodiments, the camera bracket 12 is a small-volume bracket, and an eye-tracking camera 13 is provided at one end of the camera bracket 12 away from the device frame 10. An infrared device is provided inside the eye-tracking camera 13.

[0082] In some embodiments, the camera bracket 12 can be a slender, small-volume bracket that extends outward from the device frame 10 in a certain arc. It is understood that the small-volume bracket can reduce the weight on the dog and avoid obstructing the dog's field of vision, thereby improving the accuracy of the captured eye images.

[0083] Furthermore, the eye-tracking camera 13 is typically positioned at the end of the camera mount 12 away from the device frame 10. This allows the eye-tracking camera 13 to maintain a certain shooting distance to ensure complete capture of the dog's eyes. Simultaneously, the eye-tracking camera 13 is also equipped with an infrared device, enabling it to clearly capture the dog's eye information at night, even when the dog is active. Thus, research on canine eye information is not limited to daytime but also includes the often-overlooked nighttime environment, increasing the diversity of canine eye research scenarios and improving the quality of the collected eye information.

[0084] In some embodiments, the device frame 10 is further provided with an elastic contact surface that surrounds the headband so that the device frame 10 contacts the dog when it is fixed to the dog's head.

[0085] Furthermore, the elastic contact surface can be made of sponge or rubber to enhance the comfort of dogs wearing the tracking device. This minimizes the presence of the tracking device on the dog's head, making the dog's free movement more similar to its state without the tracking device, thereby enhancing the authenticity and accuracy of the collected canine eye information.

[0086] Furthermore, the tracking device is made of waterproof and sweatproof materials to ensure that it can maintain reliability and stability even in humid environments.

[0087] In some embodiments, the device frame 10 is further provided with a panoramic camera 14, which is aligned with the dog's eye orientation to record the dog's movement trajectory and external environment information corresponding to the eye image.

[0088] Furthermore, the external environment information recorded by the panoramic camera 14 can be images of the surrounding environment taken when the dog moves to a certain place. In addition, the aforementioned GPS positioning system can also be installed in the panoramic camera 14. In this way, the movement trajectory recorded by the panoramic camera 14 can be the dog's movement footprints. By combining the movement footprints and environmental images, the range and scene of the dog's free movement can be determined, which is convenient for assisting in the analysis of the dog's eye information.

[0089] Furthermore, the panoramic camera used in this embodiment can have a resolution of 1280×720 and a frame rate of 30fps; while the eye-tracking camera can have a resolution of 640×480 and a frame rate of 30fps. This ensures that the captured images are clear and stable, and avoids the accuracy of eye tracking being affected by factors such as image blurring and shaking.

[0090] like Figure 3 As shown, Figure 3 This is an optional flowchart of a wearable head eye tracking method provided in this application embodiment, wherein the processing device is communicatively connected to the wearable head eye tracking device. The wearable head eye tracking device includes a device frame 10 and at least one eye-tracking camera 13. The device frame 10 has an opening in the middle, which is vertically continuous. The device frame 10 has downwardly protruding ear brackets 11 on both sides. The ear brackets 11 are used to mount on the dog's ears to fix the device frame 10 on the dog's head. The device frame 10 also has a protruding camera bracket 12 on the front side. The eye-tracking camera 13 is mounted on the camera bracket 12 and is used to take pictures of the dog wearing the wearable head eye tracking device to obtain eye images including the dog's eyes. Figure 3 The method may include, but is not limited to, steps S101 to S102.

[0091] Step S101: Acquire eye images sent by the wearable head eye-tracking device.

[0092] In some embodiments, a processing device connected to a tracking device can receive an eye image sent by the tracking device, and the processing device can identify and process the canine eye portion in the eye image to obtain eye information for analyzing the canine eye.

[0093] Furthermore, the tracking device can acquire real-time images of the dog's eyes and send these images to a connected processing device so that the processing device can process the images in real time.

[0094] Step S102: Identify the pupils of the canine eyes in the eye image and extract the target pupil diameter of the canine eyes.

[0095] In some embodiments, the processing device needs to identify the canine eyes in the eye image, especially the pupil, extract the target pupil diameter, and analyze the canine eye information based on the target pupil diameter.

[0096] Furthermore, the size of the target pupil diameter can reflect the impact of the surrounding environment on a dog's physiology or emotions. For example, when the surrounding fiber optic intensity is too high, a dog's pupils often constrict to protect the eyeball from damage; or, when a dog experiences emotional fluctuations, its pupil diameter often dilates; or, when a dog suffers from a subtle disease, even without abnormal changes in the surrounding environment, the dog's pupils may exhibit abnormal dilation, abnormal constriction, or asymmetry between the two pupils. In other words, identifying and processing the target pupil diameter allows for further analysis of the dog based on eye information, and obtaining high-precision eye images is a crucial step in achieving this further analysis.

[0097] Furthermore, other parts of the dog's eye can be identified and processed based on the acquired eye images, enhancing the multifaceted nature of the analysis of canine eye information. Among these, the identified and processed parts can also include the condition of the periorbital area and the sclera. Specifically, it is possible to check whether there are tears or other foreign objects around the dog's eyes and record the amount of tears and foreign objects, and to check the color of the sclera and the condition of its blood vessels. For example, the sclera is normally white, and if it is an abnormal color, it indicates that the dog's eyes may have been injured or otherwise irritated.

[0098] like Figure 4 As shown, Figure 4 yes Figure 3 A flowchart of an implementation of step S102 is provided. In some embodiments, step S102 may include steps S201 to S203.

[0099] Step S201: Input the eye image into the pre-trained pupil recognition model to extract the first diameter information of the canine eye in the eye image.

[0100] In some embodiments, the pupil recognition model is trained based on deep learning. The pupil recognition model can be any one or a combination of some of the following deep learning models: Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), Support Vector Machine (SVM), or Variational Autoencoder (VAE).

[0101] Furthermore, deep learning models can be trained using pre-defined training sets. For example, a predefined model, Residual Neural Network (ResNet), can be used as the base model, with the training set as input and TensorFlow as the training environment. The model can be trained iteratively one million times until it converges to obtain the trained model.

[0102] Furthermore, the eye-tracking camera 13 can track and record changes in the dog's eyes via video. It can acquire image frames at preset intervals and identify and process the dog's eye information within these frames. The interval can be 5 seconds to avoid wasting resources by analyzing eye images frame by frame, which would require significant computational power from the processing device. Alternatively, K-means clustering can be used to extract 400 eye images from the recorded eye videos for recognition and processing. This method of extracting eye images can also be used to generate the training set for the aforementioned pupil recognition model.

[0103] It should be noted that the interval duration can be set specifically according to the actual situation, and this application embodiment does not impose specific limitations.

[0104] Step S202: Perform image segmentation processing on the eye image according to the preset segmentation threshold to extract the second diameter information of the canine eye in the eye image.

[0105] In some embodiments, in addition to using deep learning to process eye images, image segmentation can also be used simultaneously to identify and process eye images, and to obtain the second diameter information of the canine eye in the eye image.

[0106] Furthermore, the preset segmentation threshold can be a grayscale threshold. First, the acquired eye image is processed in grayscale to obtain a grayscale image corresponding to the eye image. The grayscale value of each pixel is analyzed to determine whether it reaches the preset grayscale threshold, thereby determining the corresponding pupil region of the canine eye. Then, the target pupil diameter is determined from the pupil region. The method for determining the target pupil diameter can be an edge detection algorithm, Hough transform, or other methods. The specific method can be set according to the actual situation, and the embodiments of this application do not impose specific limitations.

[0107] Step S203: The target pupil diameter of the canine eye is obtained by fusing the first diameter information and the second diameter information.

[0108] In some embodiments, the first diameter information obtained by deep learning and the second diameter information obtained by image segmentation can be fused to obtain the final target pupil diameter. It is understood that pupil diameters obtained solely through deep learning or image processing methods contain certain errors. Fusing these two methods can minimize these errors, making the obtained target pupil diameter closer to the true value, thereby improving the accuracy of canine eye information.

[0109] It needs to be explained in more detail that although both deep learning and image segmentation methods can be used to calculate the pupil diameter of dogs, each has its own advantages and disadvantages: ① Deep learning does not require setting thresholds and parameters in specific environments. As long as the training set is broad, it can adapt to various lighting, resolution, and contrast scenarios. However, its disadvantage is that if the training set is not broad enough (i.e., scenarios with poor matching to the training set), the recognition error is extremely large; ② The advantage of image segmentation is that once the threshold is set, the pupil accuracy is very high when the lighting and contrast are stable, but the error is extremely large when the lighting changes. In this application, the embodiment uses deep learning and image segmentation methods to recognize and process the same eye image, respectively, to obtain corresponding first and second diameter information. Then, the first and second diameter information are fused, which can compensate for the error caused by using only one of the processing methods. This achieves complementary advantages and improves the accuracy of the final obtained canine target pupil diameter.

[0110] like Figure 5 As shown, Figure 5 yes Figure 4 A flowchart of an implementation of step S201 is provided. In some embodiments, step S201 may include steps S301 to S303.

[0111] Step S301: Input the eye image into the pre-trained deep learning model to obtain the pupil center point of the eye image.

[0112] In some embodiments, after inputting the eye image into a pre-trained deep learning model, the deep learning model can first perform preprocessing operations on the eye image. The preprocessing operations may include region cropping. For example, the eye image captured by the eye-tracking camera 13 may contain irrelevant background parts other than the canine eye. The deep learning model can first locate the canine eye region and then perform region cropping to obtain a preprocessed eye image that includes the pupil region. In this way, background parts irrelevant to the target can be removed, avoiding interference with subsequent processing of eye information and improving the quality of the eye image.

[0113] Furthermore, the pupil center point of the eye image can be determined from the pupil region. The pupil center point represents the position of the pupil in the eye of a dog. The position of the pupil center point in the dog's eye is of great significance for the study of canine eye information. For example, the pupil center point of a dog is usually located in the center of the iris. If a dog develops a disease, the pupil center point will deviate from the center of the iris. At this time, it can be known that the dog is very likely to have a physiological abnormality.

[0114] Step S302: Using the center point of the pupil as a reference, select any one of the outer ring feature points of the pupil in the pupil area, and determine other outer ring feature points of the pupil according to the preset interval angle. Repeat this operation until the number of outer ring feature points of the pupil reaches the preset number.

[0115] In some embodiments, besides the pupil's central point, studying the size of the pupil in dogs is also of great significance. For example, a dog's pupils dilate continuously when excited, constrict continuously when frightened, and exhibit abnormal dilation and constriction when sick. Therefore, it is necessary to determine the pupil size, i.e., the pupil diameter, as a basis for subsequent research.

[0116] Furthermore, since the pupil region is usually significantly different from other regions, after determining the center point of the pupil, a feature point can be randomly selected on the outermost side of the pupil region as the outer ring feature point of the pupil. Then, other outer ring feature points of the pupil can be determined at preset intervals.

[0117] Furthermore, the preset interval angle can be 45 degrees, thus obtaining 8 feature points on the outer ring of the pupil. The reasons for selecting 8 feature points are: ① At least 3 points are needed to perform ellipse fitting on the pupil; ② If only 3 feature points are set, the fitting accuracy of the pupil diameter will be affected if the feature point recognition error is too large; ③ 8 feature points can complement each other without having too many feature points, thus avoiding increasing the workload when manually building a deep learning feature detection model.

[0118] It should be noted that the interval angle can be set according to the actual situation. This application only describes a preferred embodiment and does not impose any specific limitations.

[0119] Step S303: Based on the pupil center point and the pupil outer ring feature points, perform a first ellipse fitting process on the eye image to obtain the first diameter information of the canine eye in the eye image.

[0120] In some embodiments, the first ellipse fitting process can be the least squares method, gradient descent method, elliptic Hough transform, or random sample consensus algorithm, etc. The first ellipse fitting process ultimately yields the first diameter information of the canine eye. The main reason for using ellipse fitting here is that when processing pupil images, due to various factors (such as lighting, angle, occlusion, etc.), the pupil may not perfectly present an ideal circular shape. Therefore, by using ellipse fitting, the shape and position of the pupil can be estimated more accurately, thereby improving the accuracy of subsequent eye analysis and processing.

[0121] Furthermore, since the pupil shape obtained after the first ellipse fitting process is usually elliptical, the major axis of the ellipse can be uniformly selected as the first diameter information, and the first diameter information can be used to reflect the size of the canine pupil.

[0122] like Figure 6 As shown, Figure 6 yes Figure 4 A flowchart of an implementation of step S202 is provided. In some embodiments, step S202 may include steps S401 to S402.

[0123] Step S401: Segment the eye image into multiple sub-images. If any sub-image is within a preset segmentation threshold, the sub-image is determined to be a set of canine eye images.

[0124] In some embodiments, the original eye image may first be preprocessed, including noise reduction, smoothing and enhancement. Then, the acquired eye image is first divided into multiple sub-image blocks using an image segmentation method. Whether the sub-image is a canine eye image set is determined by judging whether the segmented sub-image is within a preset segmentation threshold.

[0125] Step S402: Perform a second ellipse fitting process on the canine eye image set to obtain the second diameter information of the canine eyes in the eye images.

[0126] In some embodiments, corresponding feature points can be extracted from a set of canine eye images, and a second ellipse fitting process can be performed based on the feature points to obtain the second diameter information of the canine eye in the eye image.

[0127] Furthermore, the feature point extraction method for the eye image after image segmentation can be a feature point determination method similar to that in steps S301 to S302 above, or it can be other feature point determination methods such as corner detection, spot detection, or edge detection.

[0128] For example, such as Figure 7 As shown, Figure 7This is another optional flowchart of the wearable head eye tracking device provided in the embodiments of this application. After receiving the eye image, the processing device can process the eye image based on deep learning and image segmentation methods respectively to obtain the first diameter information and the second diameter information corresponding to the eye image. Then, Kalman filtering is used to remove noise from the image, and the first diameter information and the second diameter information after noise removal are fused to finally obtain a high-precision canine pupil diameter.

[0129] like Figure 8 As shown, Figure 8 yes Figure 4 A flowchart of an implementation of step S203 is provided. In some embodiments, step S203 may include steps S501 to S505.

[0130] Step S501: Calculate the first noise of the first diameter information, and determine the first state vector and the first covariance matrix corresponding to the current time based on the first noise.

[0131] In some embodiments, after obtaining the first diameter information based on deep learning and the second diameter information based on image segmentation, the two diameter information can be fused to achieve error complementarity between the two processing methods, thereby improving the accuracy of the final target pupil diameter. Here, z can be used... ROI-Seg Represents the first diameter information, and uses z dlc This indicates the second diameter information.

[0132] Furthermore, the calculated first and second diameter information usually have errors compared to the actual values. This error is usually called noise, which can be eliminated using Kalman filtering.

[0133] Furthermore, in this embodiment, noise (both the first noise and the second noise) is a random interference signal introduced during the acquisition, transmission, or processing of eye images. It is caused by various factors, such as limitations of the device itself, interference during signal transmission, and environmental influences. Therefore, it is necessary to calculate the first noise corresponding to the first diameter information and determine the first state vector and the first covariance matrix at the current moment using the first noise.

[0134] In this context, the first state vector refers to the estimated value of the current state obtained by predicting and updating the system state during the estimation process in the Kalman filter. It is a key concept in Kalman filtering and is used to represent the current state of the system. The first covariance matrix refers to the error covariance matrix of the current state obtained by predicting and updating the system state during the estimation process in the Kalman filter. It is an index describing the degree of deviation between the estimated state and the true state and reflects the uncertainty of the state estimation.

[0135] For example, firstly, the system state is estimated using a Kalman filter (KF), and smoothed data is generated after eliminating extrema. This process includes two steps: prediction and update, where the prediction step uses the following formula:

[0136]

[0137]

[0138]

[0139] p k =ap k-1 a+q (4)

[0140] In the formula, r represents observation noise, q represents process noise, and z∈R n×1 These are the observed values ​​(first diameter information and second diameter information). Let A be the posterior state estimate at time k, and let A be the state transition constant from time k-1 to time k, which can be set to 1 (assuming the current state is consistent with the previous state). k The posterior estimation error of the state estimation is recursively calculated. Specifically, to initialize KF, we will... Set z0 to z0 and p0 to 1.

[0141] Step S502: Based on the first diameter information, the first state vector, and the first covariance matrix, obtain the first state vector and the first covariance matrix corresponding to the next time step after the update.

[0142] In some embodiments, after obtaining the first diameter information, the first state vector, and the first covariance matrix at the current moment, it is necessary to predict and update the state vector and covariance matrix at the next moment based on this information. In this way, the estimation accuracy of the system state is gradually improved by continuously fusing prior estimates and observations.

[0143] For example, the update step uses the following formula:

[0144]

[0145]

[0146] p k ←(1-g k )p k (7)

[0147] Among them, g k is the Kalman gain at the current time k, used to balance the current observation with the prediction based on the previous state.

[0148] Secondly, the covariance of the observation noise significantly influences the performance of the KF fusion algorithm by affecting the fusion gain. Therefore, the estimated state is combined with the observed values ​​to dynamically recalculate the observation noise. The fusion gain is adaptively adjusted to tilt the weights towards more reliable measurement sources. The noise estimation equation is as follows:

[0149]

[0150] In the formula, ε is an infinitesimal value such that z k +ε>0.

[0151] To make the observation noise of the two measurements comparable, it can be normalized as follows:

[0152]

[0153] Step S503: Calculate the second noise of the second diameter information, and determine the second state vector and the second covariance matrix corresponding to the current time based on the second noise.

[0154] It is understandable that the calculation methods for the second noise, the second state vector, and the second covariance matrix here are similar to the calculation methods for the first noise, the first state vector, and the first covariance matrix in step S501 above, and will not be repeated here.

[0155] Step S504: Based on the second diameter information, the second state vector, and the second covariance matrix, obtain the updated second state vector and second covariance matrix for the next time step.

[0156] It is understood that the calculation method for predicting and updating the second state vector and the second covariance matrix here is similar to the calculation method for predicting and updating the first state vector and the first covariance matrix in step S502 above, and will not be repeated here.

[0157] Step S505: Determine the target pupil diameter based on the updated first state vector, first covariance matrix, second state vector, and second covariance matrix.

[0158] In some embodiments, diameter information obtained by two image processing methods and their associated data volume are fused to compensate for the errors caused by a single image processing method, which leads to low accuracy of the obtained target pupil diameter.

[0159] Furthermore, the updated first state vector and first covariance matrix can be predicted and updated again until the preset update stopping condition is reached, so as to obtain the first diameter information used to calculate the final target pupil diameter. The same applies to the second diameter information, which will not be elaborated here.

[0160] Furthermore, the update stopping condition can be that the first state vector and / or the first covariance matrix are less than a certain threshold, or it can be that the prediction update stops when a specified number of iterations is reached.

[0161] like Figure 9 As shown, Figure 9 yes Figure 8 A flowchart of an implementation of step S505 is provided. In some embodiments, step S505 may include steps S601 to S604.

[0162] Step S601: Calculate the target covariance at the current time based on the first noise and the second noise.

[0163] In some embodiments, the fusion phase further includes two steps: updating and prediction. The formula for the prediction step is as follows:

[0164]

[0165]

[0166]

[0167] P k =AP k-1 A T +Q (13)

[0168] Among them, z ROI-Seg and z dlc These are two independent pupil measurements obtained using image segmentation and deep learning methods, respectively. and These represent the observation noise estimated in the pre-fusion stage; R k Let be the covariance matrix between each pair of measurements at the current time k; Q is the covariance of the process noise. The posterior state estimate at time k; A is the state transition constant from time k-1 to time k, which can also be set to 1; P k Let be the posterior covariance matrix of the state estimated at time k, and, during initialization, we can set . for P0 is 1.

[0169] Step S602: Based on the first state vector and the second state vector, calculate the target state vector corresponding to the current time, and based on the first covariance matrix and the second covariance matrix, calculate the target covariance matrix corresponding to the current time.

[0170] In some embodiments, P k-1 represents the covariance matrix obtained at the previous time step. In this context, it represents the first diameter information and the second diameter information obtained from deep learning and image segmentation methods, respectively. By fusing the first diameter information and the second diameter information and updating them for prediction, the target covariance matrix corresponding to the current time step can be obtained.

[0171] Step S603: Based on the target covariance, obtain the updated target covariance and target covariance matrix for the next time step;

[0172] In some embodiments, after obtaining the target covariance matrix corresponding to the current time, the target covariance matrix at the current time can be used as the basis for updating the next time according to the above equation (13), and the target covariance matrix corresponding to the next time can be determined by summing it with the target covariance.

[0173] For example, after obtaining the target covariance matrix, the target covariance matrix for the next time step can be updated. Specifically, the update step uses the following formula:

[0174]

[0175]

[0176]

[0177]

[0178] P k ←(1-G k C)P k (18)

[0179] Among them, G k It is the Kalman fusion gain at the current time k, which is an adaptive weight used to adjust the weight of the measurement source on the target pupil diameter of the final fusion output.

[0180] Step S604: Determine the pupil weight value based on the proportion of the target covariance matrix, adjust the target state vector based on the pupil weight value, and determine the target pupil diameter based on the adjusted target state vector.

[0181] In some embodiments, as shown in equation (16) above, the Kalman fusion gain, i.e. the pupil weight value, can be determined by the proportion of the target covariance matrix, and as shown in equation (17) above, the obtained target state vector is adjusted by the pupil weight value, and the adjusted target state vector is determined to be the target pupil diameter.

[0182] In some embodiments, the wearable head eye-tracking device further includes a panoramic camera 14, wherein the panoramic camera 14 is aligned with the dog's eye orientation to record the dog's movement trajectory and external environment information corresponding to the eye images; the panoramic camera 14 is used to send the collected eye images and external environment information to a processing device so that the processing device processes the eye images and external environment information to obtain image display information for display.

[0183] Furthermore, the processing device that receives the eye images and external environment information sent by the panoramic camera 14 can be the same processing device that identifies and processes the canine pupils; or, a separate processing device can be selected to specifically display the panoramic images captured by the panoramic camera 14. This facilitates the comparison of images of locations the dog has passed through with the eye images captured by the eye-tracking camera 13, providing a more comprehensive information basis for the study of canine eye information.

[0184] Furthermore, in the embodiments of this application, both the eye-tracking camera and the panoramic camera record in real time. Moreover, the eye images recorded by the eye-tracking camera and the external environment information recorded by the panoramic camera can be sent to the processing device in real time, so that the processing device can perform data processing in a real-time data stream manner. Thus, the real-time performance of eye image acquisition and processing is high and the processing speed is fast.

[0185] This application also provides a processing device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned wearable head eye-tracking device and the aforementioned wearable head eye-tracking method. This processing device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0186] like Figure 10 As shown, Figure 10 This is a schematic diagram of the hardware structure of the processing device provided in an embodiment of this application. The processing device includes:

[0187] The processor 701 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0188] The memory 702 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 702 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 702 and is called and executed by the processor 701 to execute the wearable head eye-tracking device of the embodiments of this application.

[0189] The input / output interface 703 is used to implement information input and output;

[0190] The communication interface 704 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0191] Bus 705 transmits information between various components of the device (e.g., processor 701, memory 702, input / output interface 703, and communication interface 704);

[0192] The processor 701, memory 702, input / output interface 703, and communication interface 704 are connected to each other within the device via bus 705.

[0193] Furthermore, this application also provides a wearable head eye-tracking system, which includes a wearable head eye-tracking device and a processing device. The wearable head eye-tracking device included in the wearable head eye-tracking system is substantially the same as the specific embodiments of the wearable head eye-tracking devices described above, and will not be repeated here. While meeting the requirements of the embodiments of this application, the wearable head eye-tracking system may also be equipped with other functional modules to achieve other related functions.

[0194] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0195] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0196] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0197] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0198] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0199] It should be understood that in this application, "at least one" and "several" refer to one or more, and "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0200] In the embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0201] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0202] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0203] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0204] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A wearable head eye-tracking method, characterized in that, The device is used in a processing device that is communicatively connected to a wearable head-eye tracking device. The wearable head-eye tracking device includes a device frame and at least one eye-tracking camera. The device frame has an opening in the middle that runs vertically through the top and bottom. The device frame has downwardly protruding ear supports on both sides. The ear supports are used to mount on the dog's ears to fix the device frame to the dog's head. The device frame also has a protruding camera support on the front side. The eye-tracking camera is mounted on the camera support and is used to photograph the dog wearing the wearable head-eye tracking device to obtain an eye image including the dog's eyes. The method includes: Acquire the eye images sent by the wearable head eye-tracking device; The pupils of canine eyes in the eye images are identified, and the target pupil diameter of the canine eyes is extracted. The step of identifying the pupil of a canine eye in the eye image and extracting the target pupil diameter of the canine eye includes: The eye image is input into a pre-trained pupil recognition model to extract the first diameter information of the canine eye in the eye image; The eye image is segmented according to a preset segmentation threshold to extract the second diameter information of the canine eye in the eye image; The target pupil diameter of a canine eye is obtained by fusing the first diameter information and the second diameter information. The step of fusing the first diameter information and the second diameter information to obtain the target pupil diameter of a canine eye includes: Calculate the first noise of the first diameter information, and determine the first state vector and the first covariance matrix corresponding to the current time based on the first noise; Based on the first diameter information, the first state vector, and the first covariance matrix, the first state vector and the first covariance matrix corresponding to the next time step after the update are obtained. Calculate the second noise of the second diameter information, and determine the second state vector and the second covariance matrix corresponding to the current time based on the second noise; Based on the second diameter information, the second state vector, and the second covariance matrix, the updated second state vector and the second covariance matrix for the next time step are obtained. The target pupil diameter is determined based on the updated first covariance matrix and the second covariance matrix.

2. The wearable head eye-tracking method according to claim 1, characterized in that, The step of inputting the eye image into a pre-trained pupil recognition model to extract the first diameter information of the canine eye in the eye image includes: The eye image is input into a pre-trained deep learning model to obtain the pupil center point of the eye image; Using the center point of the pupil as a reference, select any one of the outer ring feature points of the pupil in the pupil area, and determine other outer ring feature points of the pupil according to a preset interval angle. Repeat this operation until the number of outer ring feature points of the pupil reaches a preset number. Based on the pupil center point and the pupil outer ring feature points, the eye image is subjected to a first ellipse fitting process to obtain the first diameter information of the canine eye in the eye image.

3. The wearable head eye-tracking method according to claim 2, characterized in that, The step of performing image segmentation processing on the eye image according to a preset segmentation threshold to extract the second diameter information of the canine eye in the eye image includes: The eye image is segmented into multiple sub-images. If any sub-image is within a preset segmentation threshold, the sub-image is determined to be a set of canine eye images. A second ellipse fitting process is performed on the set of canine eye images to obtain the second diameter information of the canine eyes in the eye images.

4. The wearable head eye-tracking method according to claim 1, characterized in that, Determining the target pupil diameter based on the updated first covariance matrix and second covariance matrix includes: Based on the first noise and the second noise, the target covariance at the current time is calculated; Based on the first state vector and the second state vector, the target state vector corresponding to the current time is calculated, and based on the first covariance matrix and the second covariance matrix, the target covariance matrix corresponding to the current time is calculated. Based on the target covariance, the updated target covariance and target covariance matrix are obtained at the next time step; The pupil weight value is determined based on the proportion of the target covariance matrix, and the target state vector is adjusted based on the pupil weight value. The target pupil diameter is then determined based on the adjusted target state vector.

5. The wearable head eye-tracking method according to claim 4, characterized in that, The wearable head eye tracking device also includes a panoramic camera, wherein the panoramic camera is aligned with the dog's eye orientation to record the dog's movement trajectory and the external environment information corresponding to the eye image; The panoramic camera is used to send the collected eye images and external environment information to the processing device, so that the processing device can process the eye images and external environment information to obtain image display information for display.

6. A wearable head eye-tracking device, characterized in that, The method for implementing the wearable head eye tracking method according to any one of claims 1 to 5 includes: The device frame has an opening in the middle that runs vertically through the frame. On both sides of the device frame are downwardly protruding ear supports that are used to mount the dog's ears to fix the device frame to the dog's head. The front of the device frame also has a protruding camera support. At least one eye-tracking camera is mounted on the camera mount and is used to photograph dogs wearing the wearable head-eye tracking device to obtain eye images including the dog's eyes. The eye image is sent to a processing device that is communicatively connected to the wearable head eye-tracking device, so that the processing device can identify the pupil of the canine eye in the eye image and extract the target pupil diameter of the canine eye.

7. The wearable head eye-tracking device according to claim 6, characterized in that, The device's outer frame is provided with an adjustable headband, which is arranged around the opening. The headband is provided with a fixing component for fixing the adjustable size of the headband.

8. The wearable head eye-tracking device according to claim 7, characterized in that, The camera bracket is a small-volume bracket, and the eye-tracking camera is installed at one end of the camera bracket away from the outer frame of the device. The eye-tracking camera is equipped with an infrared device.

9. The wearable head eye-tracking device according to claim 8, characterized in that, The device frame is also provided with an elastic contact surface, which surrounds the headband so that the device frame contacts the dog when it is fixed to the dog's head.

10. The wearable head eye-tracking device according to claim 9, characterized in that, The device frame is also equipped with a panoramic camera, which is aligned with the dog's eye orientation to record the dog's movement trajectory and the external environment information corresponding to the eye image.

11. A processing apparatus, characterized in that, The processing device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the wearable head eye-tracking method according to any one of claims 1 to 5.

12. A wearable head eye-tracking system, characterized in that, It includes the wearable head eye-tracking device according to any one of claims 6 to 10, and the processing device according to claim 11.

Images