Visual correlation head position detection method, system and equipment based on attitude information and medium

Abstract

The invention discloses a vision-related head position detection method, system and device based on posture information and a medium, and belongs to the technical field of medical instruments.The method comprises the steps that after it is confirmed that the head of a patient is kept at a normal position, head posture information of the patient is collected through a posture sensor; collecting head posture information of the patient through a posture sensor in the process that the patient watches a sighting mark or a related visual task; a head posture curve is constructed in real time according to time domain distribution, and data generated by head adjustment activities are eliminated according to curve changes; and calculating a difference value between the first space head posture information and the second space head posture information to obtain a detection result of the abnormal head position, and outputting the detection result. The problems that when an existing abnormal head position detection method is used for detection, head three-dimensional posture information cannot be accurately quantified at the same time, continuous dynamic posture information cannot be detected, and head posture information during a specific visual task cannot be detected are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical devices, and particularly relates to a visual-related head position detection method, system, device and medium based on attitude information. Background Art

[0002] An abnormal head position (AHP), also known as torticollis in medicine, is a common type of disease in pediatrics, ophthalmology, head and neck spine, and neurology. In ophthalmology, abnormal head positions are mostly caused by paralytic strabismus, nystagmus, and refractive errors. Patients adopt abnormal head positions to maintain or improve monocular or binocular visual function and eliminate diplopia. The severity of an abnormal head position is an important factor and indicator for eye disease screening, surgical intervention, and efficacy evaluation. Therefore, the accurate measurement of an abnormal head position is a very important part of patient evaluation. An abnormal head position is generally described as the rotation angles of the head around three orthogonal axes, such as facial rotation, chin elevation or depression, and head tilt.

[0003] Existing methods for measuring abnormal head positions include photography, laser projection, magnetic force and gravimeter methods, all of which have obvious defects: they cannot measure the attitudes of the head in three axial directions simultaneously, cannot be measured in specific visual tasks, cannot perform dynamic measurement and analysis, the measurement accuracy and repeatability are not high, the devices are too bulky, and wearing them for measurement interferes with the natural attitude of the subject's head. With the development of semiconductor technology and electronic technology, micro attitude sensors that use the fusion detection of accelerometers and gyroscopes have emerged, which can obtain the three-axis attitude values of the carrier in real time with high precision. However, an abnormal head position is a change in attitude relative to the normal head position. The process of a patient transitioning from a normal head position to an abnormal head position is a process and is induced in specific visual tasks; moreover, during the measurement process, the patient's head and body cannot remain fixed for a long time and there will always be random movements; some subjects also have periodic head vibrations. Therefore, when using such sensors for abnormal head position detection, a specific visual task environment is required, relevant methods, programs, and processes need to be designed, and the software and hardware media need to follow these specific processes and minimize the additional interference to the subject as much as possible in order to achieve high-precision and high-accuracy measurement and realize the storage, analysis, and reporting of measurement data. Summary of the Invention

[0004] In order to overcome the defects existing in applying the prior art to specific scenarios, the present invention provides a visual-related head position detection method, system, device and medium based on attitude information to solve the above problems.

[0005] The technical solution adopted by the present invention to solve its technical problems is: a visual-related head position detection method based on attitude information, comprising the following steps: S1: After confirming that the patient's head remains in a normal position, the patient's head posture information is collected through the posture sensor; S2: When the patient is looking at the sight mark or performing a specific visual task, the patient's head posture information is collected through the posture sensor; S3: After the measurement starts, the head posture information obtained in step S1 and the head posture information obtained in step S2 are constructed in real time according to the distribution in the time domain to form a head posture curve that is continuous in the time domain, and the data generated by the head adjustment activity is eliminated according to the curve change, so as to screen out the first spatial head posture information corresponding to when the patient's head is kept in a normal position and the second spatial head posture information corresponding to after the patient fixates on the sight mark or performs a specific visual task in the head posture curve; S4: Calculate the difference between the first spatial head posture information and the second spatial head posture information to obtain an abnormal head position detection result; S5: Output the detection result of abnormal head position.

[0006] It is worth noting that, in step S3, the baseline segment of the head posture curve in each round of acquisition process is used as the band in which the patient's head remains in a normal position, the peak segment or the trough segment of the head posture curve in each round of acquisition process is used as the band after the patient tilts his head, and the band in which the head posture curve in each round of acquisition process transitions from the baseline segment to the peak segment or from the baseline segment to the trough segment is used as the data generated by the head adjustment activity.

[0007] Preferably, in step S1 and step S2, the head posture information includes a collected pitch value, a collected roll value, and a collected yaw value; before starting the measurement, the collection time of the posture sensor is reset to 0, and during the measurement, a time node corresponding to each collected head posture information in the current task is assigned.

[0008] Optionally, in step S3, the head posture curve includes a pitch curve constructed by collecting pitch values ​​according to the distribution in the time domain, a roll curve constructed by collecting roll values ​​according to the distribution in the time domain, and a yaw curve constructed by collecting yaw values ​​according to the distribution in the time domain; The first spatial head posture information includes a starting pitch value, a starting roll value, and a starting yaw value; The second spatial head posture information includes a measured pitch value, a measured roll value, and a measured yaw value; In step S4, the difference between the average value of the starting pitch value and the average value of the measured pitch value is calculated as the pitch detection value, the difference between the average value of the starting roll value and the average value of the measured roll value is calculated as the roll detection value, and the difference between the average value of the starting yaw value and the average value of the measured yaw value is calculated as the yaw detection value. Finally, the pitch detection value, the roll detection value and the yaw detection value are used as the detection results of the abnormal head position.

[0009] Specifically, the Y-axis of the attitude sensor is the vertical direction, and the deflection variable corresponding to the attitude sensor is the yaw value; the Z-axis of the attitude sensor is the front-back direction, and the deflection variable corresponding to the attitude sensor is the roll value; the X-axis of the attitude sensor is the left-right direction, and the deflection variable corresponding to the attitude sensor is the pitch value.

[0010] It should be noted that before performing the step S2, first confirm that the patient is in a sitting position, and then confirm that the attitude sensor is fixed on the patient's head so that the attitude sensor keeps moving synchronously with the patient's head.

[0011] Preferably, in the steps S1 and S2, during the process of the attitude sensor collecting the head attitude information, the head of the patient is synchronously recorded, and the recording time is matched with the collection time of the head attitude information.

[0012] It should be noted that an abnormal head position detection system based on an attitude sensor includes: The first information collection module: used to collect the head attitude information of the patient through the attitude sensor after confirming that the patient's head remains in the normal position; The second information collection module: used to collect the head attitude information of the patient through the attitude sensor during the process of the patient gazing at the visual target or performing a specific visual task; The curve construction and processing module: used to, after the measurement starts, construct a continuous head attitude curve in the time domain in real time from the head attitude information obtained from the first information collection module and the head attitude information obtained from the second information collection module according to the time domain distribution, and eliminate the data generated by the head adjustment activities according to the curve change, so as to screen out the first spatial head attitude information corresponding to the patient's head remaining in the normal position and the second spatial head attitude information corresponding to the patient gazing at the visual target or performing a specific visual task in the head attitude curve; The comparison and calculation module: calculates the difference between the first spatial head attitude information and the second spatial head attitude information to obtain the detection result of the abnormal head position; The output module: outputs the detection result of the abnormal head position.

[0013] A computer device includes: a memory and a processor; the memory stores a computer program, and when the processor executes the computer program, it implements the steps of the vision-related head position detection method based on attitude information according to any one of claims 1 to 7.

[0014] A computer-readable storage medium stores a computer program thereon, and when the computer program is executed by a processor, it implements the steps of the vision-related head position detection method based on attitude information according to any one of claims 1 to 7.

[0015] The beneficial effects of the present invention are as follows: In the visual-related head position detection method based on attitude information, when the head attitude information of the patient is collected by the attitude sensor, whether it is the data at the tilted head position, the data at the upright position, or the data during the head turning process, they will all be recorded first. After forming the head attitude curve, the head attitude curve is processed through step S3, so that only the wave bands at the tilted head position and the wave bands at the upright position are left on the processed curve. By comparing the two wave bands, the data of the head position that can reflect abnormalities can be obtained, and thus the detection result can be obtained. During this process, the patient only needs to face the detection according to his own natural reaction, which facilitates the doctor's detection operation on the patient. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a flowchart of the visual-related head position detection method based on attitude information in an embodiment of the present invention; Figure 2 is a schematic diagram of the coordinate system construction of the attitude sensor in an embodiment of the present invention; Figure 3 is a photo obtained by shooting with a video recorder in an embodiment of the present invention; Figure 4 is a schematic diagram of the head attitude curve in an embodiment of the present invention; Figure 5 is a schematic diagram when the patient's head remains in the normal position in an embodiment of the present invention; Figure 6 is a schematic diagram when the patient's head is at the tilted head position in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] The following further describes the specific embodiments of the present invention with reference to the accompanying drawings. It should be noted here that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation to the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0018] As Figures 1-6 shown, a visual-related head position detection method based on attitude information includes the following steps: S1: After confirming that the patient's head remains in the normal position, collect the head attitude information of the patient through the attitude sensor; in this embodiment, the normal position means the position where the patient's head is visually confirmed by the family members and the doctor together without being tilted, that is Figure 2 shown: The Y-axis is perpendicular to the ground, and the Z-axis points to the fixation target; S2: During the process of the patient gazing at the visual target or performing a specific visual task, the head pose information of the patient is collected by the pose sensor; in this embodiment, the specific visual task is that the examinee gazes at pictures, animations, and videos corresponding to their visual acuity level, and at this time, foveal fixation is used. As Figure 2 shown, when performing step S1 and step S2, the pose sensor is worn on the patient's head; the Y-axis of the pose sensor is the vertical direction, and the corresponding deflection variable of the pose sensor is the yaw value; the Z-axis of the pose sensor is the front-back direction, and the corresponding deflection variable of the pose sensor is the roll value; the X-axis of the pose sensor is the left-right direction, and the corresponding deflection variable of the pose sensor is the pitch value; since the Y-axis of the pose sensor is always in the direction of gravity perpendicular to the ground, which is consistent with the Y-axis direction of the patient's head pose coordinate system, that is, the direction from the patient's head top to the ground. The front direction of the patient's face is the Z-axis direction of the head pose coordinate system, and the left and right ear sides of the patient are the X-axis direction of the head pose coordinate system. Therefore, when marking the data acquisition, it is necessary to confirm the specific meaning and positive and negative sign directions of the three-axis measurement data of the pose sensor, so as to facilitate interpreting the head pose of the acquired data. The acquired pitch value represents the amount of the patient's chin moving up or down, the acquired roll value represents the amount of the patient's head tilting, and the acquired yaw value represents the amount of the patient's face turning. The model of the pose sensor is LPMS-B2.

[0019] In step S1 and step S2, the head pose information includes the acquired pitch value, the acquired roll value, and the acquired yaw value; before starting the measurement, the acquisition time of the pose sensor is reset to 0, and during the measurement, the corresponding time node is assigned to each acquired head pose information in the current task. That is, after resetting the acquisition time of the pose sensor to 0, starting from the time node of 0, the corresponding time node is assigned to each acquired head pose information. For example, if the acquisition interval is 1 s, the time node of the first acquired head pose information is 1 s, the time node of the second acquired head pose information is 2 s, and so on; Before performing step S1, it will first be confirmed that the patient is in a sitting position, and then it will be confirmed that the pose sensor is fixed on the patient's head so that the pose sensor can keep moving synchronously with the patient's head. In this embodiment, the pose sensor is fixed on a disposable medical cap, and then the pose sensor is fixed on the patient's head through the disposable medical cap to achieve synchronous movement of the pose sensor and the patient's head.

[0020] S3: After the measurement starts, the head posture information obtained in step S1 and the head posture information obtained in step S2 are constructed in real time according to the distribution in the time domain to form a head posture curve that is continuous in the time domain, and the data generated by the head adjustment activity is eliminated according to the curve change, so as to screen out the first spatial head posture information corresponding to when the patient's head is kept in a normal position (i.e., the head posture information of the normal posture) and the second spatial head posture information corresponding to when the patient fixates on the sight mark or performs a specific visual task (i.e., the head posture information of the abnormal posture) in the head posture curve; Figure 4 As shown, in this embodiment, since the head posture information obtained in step S2 is collected first, and the head posture information obtained in step S3 is collected later, in the obtained head posture curve, the second space head posture information obtained in each round of collection process will be after the first space head posture information, and there will be a curve segment connecting the first space head posture information and the second space head posture information. This curve segment is the process of the patient's head turning during the process of looking at the sight mark or a specific visual task, and is the data generated by the head posture adjustment activity. In order not to affect the calculation between the first space head posture information and the second space head posture information in the subsequent steps, it needs to be eliminated. In this embodiment, each detection will repeat steps S2 and S3 three times to perform three rounds of collection, and obtain three head posture curves to ensure the accuracy of the data. In this embodiment, abnormal head movements are identified by manually or AI monitoring and comparing the morphological changes of the head video images and curves. Specifically, the sight mark or related visual task that the patient needs to look at is set at a distance of 5 meters from the patient, and the sight mark or related visual task that the patient needs to look at is level with the patient's eyes. Then the patient is asked to look at the sight mark or related visual task, and the patient is induced to tilt his head position during the process of looking at the sight mark or related visual task. For children who are unable to cooperate with vision tests, brightly colored pictures that they can see clearly are used as sight marks.

[0021] In step S3, the head posture curve includes a pitch curve constructed by collecting pitch values ​​according to the distribution in the time domain, a roll curve constructed by collecting roll values ​​according to the distribution in the time domain, and a yaw curve constructed by collecting yaw values ​​according to the distribution in the time domain; The first spatial head posture information includes a starting pitch value, a starting roll value, and a starting yaw value; The second spatial head posture information includes a measured pitch value, a measured roll value, and a measured yaw value.

[0022] In the steps S1 and S2, during the process of the attitude sensor collecting the head attitude information, the head of the patient is synchronously recorded, and the recording time is matched with the collection time of the head attitude information. In this embodiment, preferably, a video recorder is set at the head attitude where the patient is looking, and the lens of the video recorder is directed towards the head of the patient for recording. The specific operation is to first synchronize the system time of the attitude sensor with the system time of the video recorder, and record the video synchronously when collecting information through the attitude sensor, so that the recording time is matched with the collection time of the head attitude information. When constructing the head attitude curve, the time domain of the head attitude curve will be matched with the system time of the recording, so that the corresponding head attitude information can be found on the head attitude curve through the recording, which can more intuitively display the state of the patient and is convenient for subsequent tracing. The video obtained by the video recorder can be processed to obtain photos at the corresponding time points, and after adding angle marks, it can be obtained as shown in Figure 3 The figure shown in Figure 3 . In

[0023] , the left photo can reflect the roll amount of the patient, that is, the amount of tilt of the patient's head, the middle photo can reflect the pitch amount of the patient, that is, the amount of the patient's chin moving up or down, and the right photo can reflect the yaw amount of the patient, that is, the amount of rotation of the patient's face. Figure 4 The curves shown in Figure 4Among them, "baseline" represents the frontal position, "peak" represents the tilted head, "Roll" represents the measured roll value, "Pitch" represents the measured pitch value, and "Yaw" represents the measured yaw value. Each round of acquisition corresponds to a head pose curve, that is, a base segment, a transition segment, and a peak segment, or a base segment, a transition segment, and a trough segment as a head pose curve.

[0024] When the measurement ends, the system displays the continuous changes of the rotation angles of the patient's head in three axes (representing the acquired pitch value, the acquired roll value, and the acquired yaw value respectively) in the form of the curves shown as Figure 4 (that is, three head pose curves). The doctor or the measurer judges the integrity of the measurement result according to the three head pose curves. In this embodiment, the three head pose curves should include three peak segments and three base segments (because the measurement is repeated three rounds), and each part should last for a certain time length (the maintenance time of the measurement), and quantitative analysis is performed on those with qualified image quality. The doctor can move the marking line on the touch screen with a finger to mark the stable base segment corresponding to the frontal position in the three rounds and the peak segment or trough segment corresponding to the tilted head in the curve.

[0025] S4: Calculate the difference between the first spatial head pose information and the second spatial head pose information to obtain the detection result of the abnormal head position; in the step S4, calculate the difference between the average value of the starting pitch value and the average value of the measured pitch value as the pitch detection value, calculate the difference between the average value of the starting roll value and the average value of the measured roll value as the roll detection value, calculate the difference between the average value of the starting yaw value and the average value of the measured yaw value as the yaw detection value, and finally use the pitch detection value, the roll detection value, and the yaw detection value as the detection result of the abnormal head position.

[0026] S5: Output the detection result of the abnormal head position in combination with the dynamic information such as the vibration frequency and amplitude of the head pose curve. In this embodiment, through the algorithm of the software, the vibration curve can be obtained in the head pose curve for the analysis of the frequency and amplitude. This analysis is performed after the measurement ends and does not require real-time analysis. Analyze the head pose curve. For people with nystagmus disease, their heads will vibrate involuntarily. At this time, the data curve will show periodic vibration. In this way, the frequency (times / second), amplitude (maximum amplitude, minimum amplitude, average amplitude, etc.) of the vibration can be calculated according to the analysis and description method of the general vibration curve, and the type of the vibration curve can be analyzed: impulse type, pendulum type, irregular type.

[0027] In the above-mentioned vision-related head position detection method based on attitude information, when collecting the head attitude information of the patient through the attitude sensor, whether it is the data at the tilted head position, the data at the straight head position, or the data during the head turning process, they will all be recorded first. After forming the head attitude curve, the head attitude curve is processed through step S3, so that only the wave bands at the tilted head position and the wave bands at the straight head position are left in the processed curve. By comparing the two wave bands, the data of the head position that can reflect abnormalities can be obtained, and thus the detection result can be obtained. During this process, the patient only needs to face the detection according to his or her natural reaction, which facilitates the doctor's detection operation on the patient.

[0028] Specifically, after clicking "Measurement and Analysis" on the program interface, the system will analyze the angles of the three head tilts according to the markings, and then compare them with the first spatial head attitude information collected by the first information collection module, and generate a report for saving. After saving, it can also be instantaneously synchronized and transmitted through WeChat and other means.

[0029] It should be noted that an abnormal head position detection system based on an attitude sensor includes: The first information collection module: used to collect the head attitude information of the patient through the attitude sensor after confirming that the patient's head is in the normal position; The second information collection module: used to collect the head attitude information of the patient through the attitude sensor during the process when the patient is gazing at the visual target or performing a specific visual task; The curve construction and processing module: used to, after the measurement starts, construct a head attitude curve that is continuous in the time domain in real time according to the time domain distribution of the head attitude information obtained from the first information collection module and the head attitude information obtained from the second information collection module, and eliminate the data generated by the head adjustment activities according to the curve changes, so as to screen out the first spatial head attitude information corresponding to the patient's head staying in the normal position and the second spatial head attitude information corresponding to the patient after gazing at the visual target or performing a specific visual task in the head attitude curve; The comparison and calculation module: calculates the difference between the first spatial head attitude information and the second spatial head attitude information to obtain the detection result of the abnormal head position; The output module: outputs the detection result of the abnormal head position.

[0030] It should be noted that a computer device includes: a memory and a processor; the memory stores a computer program, and when the processor executes the computer program, it implements the steps of the above-mentioned vision-related head position detection method based on attitude information.

[0031] It should be noted that a computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the steps of the above-mentioned vision-related head position detection method based on attitude information.

[0032] The embodiments of the present invention have been described in detail above in conjunction with the accompanying drawings, but the present invention is not limited to the described embodiments. For those skilled in the art, without departing from the principle and spirit of the present invention, various changes, modifications, substitutions, and variations to these embodiments still fall within the protection scope of the present invention.

Claims

1. A visual correlation head position detection method based on posture information, characterized in that: The following steps are involved: S1: After confirming that the patient's head remains in a normal position, the patient's head posture information is collected through the posture sensor; S2: When the patient is looking at the sight mark or performing a specific visual task, the patient's head posture information is collected through the posture sensor; S3: After the measurement starts, the head posture information obtained in step S1 and the head posture information obtained in step S2 are constructed in real time according to the distribution in the time domain to form a head posture curve that is continuous in the time domain, and the data generated by the head adjustment activity is eliminated according to the curve change, so as to screen out the first spatial head posture information corresponding to when the patient's head is kept in a normal position and the second spatial head posture information corresponding to after the patient fixates on the sight mark or performs a specific visual task in the head posture curve; S4: Calculate the difference between the first spatial head posture information and the second spatial head posture information to obtain an abnormal head position detection result; S5: Output the detection result of abnormal head position.

2. The method for detecting head position based on visual correlation of posture information according to claim 1, characterized in that: In step S3, the baseline segment of the head posture curve in each round of acquisition process is used as the band in which the patient's head remains in a normal position, the peak segment or the trough segment of the head posture curve in each round of acquisition process is used as the band after the patient tilts his head, and the band in which the head posture curve in each round of acquisition process transitions from the baseline segment to the peak segment or from the baseline segment to the trough segment is used as the data generated by the head adjustment activity.

3. The method for detecting head position based on visual correlation of posture information according to claim 2, characterized in that: In step S1 and step S2, the head posture information includes the collected pitch value, the collected roll value and the collected yaw value; before starting the measurement, the collection time of the posture sensor is reset to 0, and during the measurement, a time node corresponding to each collected head posture information in the current task is assigned.

4. The method for detecting head position based on visual correlation of posture information according to claim 3, characterized in that: In step S3, the head posture curve includes a pitch curve constructed by collecting pitch values ​​according to the distribution in the time domain, a roll curve constructed by collecting roll values ​​according to the distribution in the time domain, and a yaw curve constructed by collecting yaw values ​​according to the distribution in the time domain; The first spatial head posture information includes a starting pitch value, a starting roll value, and a starting yaw value; The second spatial head posture information includes a measured pitch value, a measured roll value, and a measured yaw value; In step S4, the difference between the average value of the starting pitch value and the average value of the measured pitch value is calculated as the pitch detection value, the difference between the average value of the starting roll value and the average value of the measured roll value is calculated as the roll detection value, and the difference between the average value of the starting yaw value and the average value of the measured yaw value is calculated as the yaw detection value. Finally, the pitch detection value, the roll detection value and the yaw detection value are used as the detection results of the abnormal head position.

5. The method for detecting head position based on visual correlation of posture information according to claim 4, characterized in that: When executing step S1 and step S2, the posture sensor is worn on the patient's head; the Y-axis of the posture sensor is the vertical direction, and the deflection variable of the corresponding posture sensor is the yaw value; the Z-axis of the posture sensor is the front-to-back direction, and the deflection variable of the corresponding posture sensor is the roll value; the X-axis of the posture sensor is the left-right direction, and the deflection variable of the corresponding posture sensor is the pitch value.

6. The method for detecting head position based on visual correlation of posture information according to claim 5, characterized in that: Before executing step S2, it is first confirmed that the patient is in a sitting position, and then it is confirmed that the posture sensor is fixed to the patient's head so that the posture sensor keeps moving synchronously with the patient's head.

7. The method for detecting head position based on visual correlation of posture information according to claim 6, characterized in that: In the step S1 and step S2, while the posture sensor is collecting the head posture information, the patient's head is synchronously recorded, and the recording time is matched with the collection time of the head posture information.

8. An abnormal head position detection system based on a posture sensor, characterized in that: include: The first information collection module is used to collect the patient's head posture information through the posture sensor after confirming that the patient's head remains in a normal position; The second information collection module is used to collect the patient's head posture information through the posture sensor when the patient is looking at the sight mark or performing a specific visual task; Curve construction processing module: used for constructing a head space head posture curve that is continuous in time domain in real time according to the distribution in time domain based on the head posture information obtained from the first information acquisition module and the head posture information obtained from the second information acquisition module after the measurement starts, and eliminating the data generated by the head adjustment activity according to the curve change, so as to screen out the first space head posture information corresponding to when the patient's head is kept in a normal position and the second space head posture information corresponding to when the patient fixates on the sight mark or performs a specific visual task in the head posture curve; Comparison calculation module: calculates the difference between the first spatial head posture information and the second spatial head posture information to obtain the abnormal head position detection result; Output module: outputs the detection results of abnormal head positions.

9. A computer device comprising: A memory and a processor; the memory stores a computer program, wherein the processor implements the steps of the visual-related head position detection method based on posture information according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the visual correlation head position detection method based on posture information according to any one of claims 1 to 7 are implemented.

Images