CN117860191A - Method for fully-automatically detecting eyeball center for fixing head of tonometer - Google Patents

Abstract

The invention discloses a method for fully-automatically detecting the center of an eyeball by fixing the head of a tonometer, which comprises the following steps: step one, a voice module and a screen on a tonometer are demonstrated to prompt a patient to place a chin on a mandibular support and press a forehead close to a forehead cushion; step two, detecting the weight by a weight sensor; step three, detecting whether skin contact exists by a capacitance sensor in the forehead cushion; step four, after the head position is accurately adjusted, the wide-angle camera firstly recognizes the face, and then the approximate position of human eyes is judged; converting the image coordinate system into a physical coordinate system, and moving the triaxial mobile platform to the vicinity of the target; and step six, the binocular camera recognizes eyeballs, the triaxial moving platform performs eye ball alignment operation, when the images seen by the binocular camera are consistent, the center of the eyeballs are aligned, the triaxial moving platform stops moving, and an intraocular pressure detection flow is entered. The beneficial effects of the invention are as follows: the purpose of improving the stability of the intraocular pressure measurement result is achieved.

Description

Method for fully-automatically detecting eyeball center for fixing head of tonometer

Technical Field

The invention relates to the technical field related to non-contact tonometers, in particular to a method for fixing the head of a tonometer and fully automatically detecting the center of an eyeball.

Background

At present, most people in society have eye diseases, such as myopia, glaucoma, achromatopsia, astigmatism, color weakness and the like, and intraocular pressure is taken as an important reference index of eye health, so that intraocular pressure detection is almost a necessary detection item of all ophthalmic hospitals.

Chinese patent No. CN214965420U, grant bulletin day 2021, 12 months 03 days, disclose a tonometer operation frame, including arc bracing piece, horizontal support plate, face support groove, telescopic link and limiting plate, horizontal support plate is horizontal support plate, face support groove is notch ascending arc groove, face support groove's lower terminal surface is fixed the middle part of horizontal support plate up end, be equipped with vertical limiting plate on the face support groove, the limiting plate passes through the telescopic link is fixed on the horizontal support plate, the telescopic link includes L type telescopic inner rod and vertical telescopic outer rod, be equipped with inner rod standing groove and inner ring arch on the vertical telescopic outer rod, L type telescopic inner rod passes through vertical rod is placed vertical telescopic outer rod in the inner rod standing groove, arc bracing piece fixed connection the lower terminal surface of horizontal support plate.

As known from the tonometer operation frame disclosed in the above patent, the current tonometers in the main stream of hospitals mainly rely on manual work to guide patients to adjust the positions of the faces in the operation frame, which is time-consuming and laborious, and has a certain influence on accuracy, so that the tonometer measurement result is unstable.

Disclosure of Invention

The invention provides a method for fully-automatic detection of eyeball center for fixing the head of a tonometer, which is beneficial to improving the stability of the tonometer and aims to overcome the defect of unstable tonometer results in the prior art.

In order to achieve the above purpose, the present invention adopts the following technical scheme:

a method for fully automatic detection of the center of an eyeball for tonometer head fixation, comprising the steps of:

step one, a voice module and a screen on a tonometer are demonstrated to prompt a patient to place a chin on a mandibular support and press a forehead close to a forehead cushion;

step two, detecting the weight by a weight sensor in the mandibular rest, and carrying out corresponding prompt and demonstration through a voice module and a screen if necessary;

step three, a capacitance sensor in the forehead cushion detects whether skin contact exists or not, and corresponding prompt and demonstration are carried out through a voice module and a screen if necessary;

step four, after the head position is accurately adjusted, the wide-angle camera firstly recognizes the face, and then the approximate position of human eyes is judged;

converting the image coordinate system into a physical coordinate system, and moving the triaxial mobile platform to the vicinity of the target;

and step six, the binocular camera recognizes eyeballs, the triaxial moving platform performs eyeball alignment operation, when the images seen by the binocular camera are consistent, the center of the eyeballs are aligned, the triaxial moving platform stops moving, and the next intraocular pressure detection flow can be entered.

The weight sensor is used for detecting whether the chin is placed on the mandibular support, and the capacitance sensor is used for detecting whether the forehead clings to the forehead cushion. The voice module and the screen can prompt and guide simultaneously during detection. The machine vision part comprises a binocular camera and a wide-angle camera, the wide-angle camera firstly recognizes a human face and then roughly determines the approximate positions of eyeballs, the binocular camera part is used for detecting the accurate positions of the left eye and the right eye and searching the center of the eyeballs, so that manual guidance adjustment is replaced, the purpose of improving stability of intraocular pressure measurement results is achieved, the center of the eyeballs of a patient is detected fully automatically, manual alignment is not needed, and the whole process of the tonometer is automatically prompted and aligned.

Preferably, in the second step, when the weight detected by the weight sensor is lower than the set pressure threshold, the voice module and the screen will prompt and demonstrate correspondingly. The prompting content of the voice is that' for the accuracy of detection, please put the chin on the mandible support and keep the chin, and correspondingly demonstrate through the screen, so that the patient can quickly and accurately fix the head, the work efficiency is improved, and the accuracy of detection and the stability of the measurement result are improved.

Preferably, the set pressure threshold is 400g. When the weight sensor receives a force in the vertical direction, the voltage will increase linearly with the force, so we set the reasonable pressure threshold through experiments.

Preferably, in the third step, when the capacitive sensor does not detect skin, the voice module and the screen will perform corresponding prompt and demonstration. The prompting content of the voice is that the forehead is clung to the forehead cushion for the accuracy of detection and kept motionless, and corresponding demonstration is carried out through a screen, so that the head of a patient can be conveniently and rapidly fixed, the work efficiency is improved, and the accuracy of detection and the stability of a measurement result are improved.

Preferably, the tonometer comprises a detection main body, a head placement area is arranged on the side edge of the detection main body, a forehead cushion and a mandibular support are arranged in the head placement area, the forehead cushion is fixedly connected with the top of the head placement area, a capacitance sensor is arranged in the forehead cushion, a weight sensor is arranged at the bottom of the head placement area, the mandibular support acts on the weight sensor and is detachably connected with the bottom of the head placement area, a control main board, a triaxial mobile platform and a voice module are detachably connected in the detection main body, a wide-angle camera and a binocular camera which are corresponding to the head placement area are arranged on the triaxial mobile platform, a screen is connected on the detection main body, and the capacitance sensor, the weight sensor, the triaxial mobile platform, the voice module, the wide-angle camera, the binocular camera and the screen are all electrically connected with the control main board. The patient uses non-contact automated inspection tonometer, and voice module can play and be used for prompting patient's operation, and simultaneously, the screen can carry out corresponding animation demonstration. According to the prompt, the patient places the chin on the chin rest and places the forehead close to the forehead cushion. The weight sensor detects weight, and when the weight is lower than the pressure threshold value, the voice and the screen can correspondingly prompt and demonstrate. The prompting content of the voice is "please put the chin on the mandibular rest for the accuracy of detection, and keep still". The capacitive sensor detects whether there is skin contact, and changes from low to high when there is skin contact. If the capacitive sensor does not detect skin, the voice and the screen can correspondingly prompt and demonstrate. The prompting content of the voice is 'for detecting accuracy, please attach the forehead to the forehead cushion and keep the forehead motionless'; when the sensors of the mandibular rest and the forehead cushion detect the action of a person, the tonometer can enter the next detection. The response time of the weight sensor and the capacitance sensor is in microsecond level, and the time of the using process is not influenced. The above procedure ensures that the patient maintains a standard test posture during the test, facilitating measurement of ocular tension. The eyeball detection flow is divided into a coarse adjustment process and a fine adjustment process. The rough adjustment process is to firstly identify the face by a wide-angle and high-resolution camera (the selected camera has strong self-adaption capability and can identify human eyes under different light environments), and then judge the approximate position of the human eyes. After the rough position is found, the image coordinate system is converted into a physical coordinate system, and the three-axis moving platform of the tonometer moves to the vicinity of the target. At this time, the binocular camera performs fine adjustment. At this time, the triaxial moving platform is aligned with the eyeball. When the images seen by the binocular camera are consistent, the center of the eyeball is aligned, the triaxial mobile platform stops moving, and the next flow can be entered. The invention can greatly reduce the time of medical treatment for hospitals and patients and reduce the cost of people for hospitals.

Preferably, the detection main body is provided with a support column, one end of the support column is located in the detection main body, the other end of the support column is located outside the detection main body, the support column is connected with the detection main body in a vertical sliding mode, the weight sensor and the support column are located in one end outside the detection main body and are detachably connected, and the mandible support is connected with one end of the support column located outside the detection main body and acts on the weight sensor. The support column and detect from top to bottom sliding connection of main part do benefit to and carry out altitude mixture control according to patient's face size, do benefit to the improvement practicality.

Preferably, the mandible support is provided with a connecting shaft, one end of the connecting shaft is fixedly connected with the center of the mandible support, the other end of the connecting shaft is connected with a weight sensor, the connecting shaft is detachably connected with the inside of the support column, and the mandible support is positioned outside the support column.

Preferably, the detection main body is internally provided with a screw motor and a lifting table, the screw motor is electrically connected with the control main board, the screw motor is detachably connected with the detection main body, the output end of the screw motor is connected with the lifting table, the lifting table is in sliding connection with the lifting table up and down under the driving of the screw motor, and one end of the support column in the detection main body is detachably connected with the lifting table. The screw rod motor is controlled to work through the control main board, so that the accuracy and the sensitivity beneficial to the adjustment of the height of the mandibular support are improved.

Preferably, the binocular camera is an infrared camera. The binocular camera is an infrared camera, the iris is obvious under infrared light, and the recognition of the eyeball is more accurate.

Preferably, the wide-angle camera is located above the binocular camera, the mandibular support is located below the binocular camera, the binocular camera comprises a left-eye camera and a right-eye camera, and the left-eye camera and the right-eye camera are distributed in bilateral symmetry with the wide-angle camera as a center. The selected wide-angle camera has high resolution and strong self-adaption capability, and can identify human eyes in different light environments; the left eye camera is used for identifying left eyeballs, and the right eye camera is used for identifying right eyeballs.

The beneficial effects of the invention are as follows: the invention can greatly reduce the time of medical treatment for hospitals and patients and reduce the cost of personnel for hospitals; the working efficiency is improved, and the detection accuracy is improved; the height adjustment is facilitated according to the face size of the patient, and the practicability is improved.

Drawings

FIG. 1 is a schematic flow chart of the present invention;

FIG. 2 is a schematic diagram of the structure of the tonometer;

fig. 3 and 7 are partial internal structural schematic views of the tonometer;

FIG. 4 is a front view of the head rest area of FIG. 3;

FIG. 5 is a cross-sectional view of A-A of FIG. 4;

fig. 6 is an enlarged view of the structure at B in fig. 5.

In the figure: 1. the device comprises a detection main body, a head placement area, a forehead cushion, a mandibular support, a capacitive sensor, a weight sensor, a triaxial moving platform, a voice module, a wide-angle camera, a binocular camera, a screen, a support column, a connecting shaft, a screw motor, a lifting platform, a left-eye camera and a right-eye camera.

Description of the embodiments

The invention is further described below with reference to the drawings and detailed description.

In the embodiment shown in fig. 1, a method for fully automatic eye center detection for tonometer head fixation includes the steps of:

step one, a voice module 8 and a screen 11 on the tonometer are demonstrated to prompt a user to place a chin on the mandibular support 4 and press the forehead close to the forehead cushion 3;

step two, a weight sensor 6 in the mandibular rest 4 detects weight, and corresponding prompt and demonstration are carried out through a voice module 8 and a screen 11 if necessary;

step three, the capacitive sensor 5 in the forehead cushion 3 detects whether skin contact exists, and corresponding prompt and demonstration are carried out through the voice module 8 and the screen 11 if necessary;

step four, after the head position is accurately adjusted, the wide-angle camera 9 firstly recognizes the human face and then judges the approximate position of human eyes;

step five, converting the image coordinate system into a physical coordinate system, and moving the triaxial moving platform 7 to the vicinity of the target;

step six, the binocular camera 10 recognizes the eyeballs, the triaxial moving platform 7 performs the operation of aligning the eyeballs, when the images seen by the binocular camera 10 are consistent, the center of the eyeballs are aligned, the triaxial moving platform 7 stops moving, and the next intraocular pressure detection process can be entered.

In the second step, when the weight detected by the weight sensor 6 is lower than the set pressure threshold, the voice module 8 and the screen 11 will prompt and demonstrate correspondingly. The pressure threshold was set at 400g.

In step three, when the capacitive sensor 5 does not detect skin, the voice module 8 and the screen 11 will perform corresponding prompting and demonstration.

As shown in fig. 2, fig. 3, fig. 4, fig. 5 and fig. 6, the tonometer comprises a detection main body 1, a head placement area 2 is arranged on the side edge of the detection main body 1, a forehead cushion 3 and a mandibular support 4 are arranged in the head placement area 2, the forehead cushion 3 is fixedly connected with the top of the head placement area 2, a capacitance sensor 5 is arranged in the forehead cushion 3, a weight sensor 6 is arranged at the bottom of the head placement area 2, the mandibular support 4 acts on the weight sensor 6 and is detachably connected with the bottom of the head placement area 2, a control main board, a triaxial moving platform 7 and a voice module 8 are detachably connected in the detection main body 1, a wide-angle camera 9 and a binocular camera 10 which are all corresponding to the head placement area 2 are arranged on the triaxial moving platform 7, a screen 11 is connected on the detection main body 1, and the capacitance sensor 5, the weight sensor 6, the triaxial moving platform 7, the voice module 8, the wide-angle camera 9, the binocular camera 10 and the screen 11 are all electrically connected with the control main board.

As shown in fig. 5, the detection main body 1 is provided with a support column 12, one end of the support column 12 is located in the detection main body 1, the other end of the support column 12 is located outside the detection main body 1, the support column 12 is connected with the detection main body 1 in a vertically sliding manner, the weight sensor 6 is connected with the support column 12 in a detachable manner at one end of the detection main body 1, and the mandibular support 4 is connected with one end of the support column 12 located outside the detection main body 1 and acts on the weight sensor 6.

As shown in fig. 5, the mandibular support 4 is provided with a connecting shaft 13, one end of the connecting shaft 13 is fixedly connected with the center of the mandibular support 4, the other end of the connecting shaft 13 is connected with the weight sensor 6, the connecting shaft 13 is detachably connected with the inside of the support column 12, and the mandibular support 4 is positioned outside the support column 12.

As shown in fig. 7, a screw motor 14 and a lifting table 15 are arranged in the detecting main body 1, the screw motor 14 is electrically connected with the control main board, the screw motor 14 is detachably connected with the detecting main body 1, the output end of the screw motor 14 is connected with the lifting table 15, the lifting table 15 is in sliding connection up and down under the driving of the screw motor 14, and one end of the support column 12 positioned in the detecting main body 1 is detachably connected with the lifting table 15. The binocular camera 10 is an infrared camera.

As shown in fig. 3, the wide-angle camera 9 is located above the binocular camera 10, the mandibular rest 4 is located below the binocular camera 10, the binocular camera 10 includes a left-eye camera 16 and a right-eye camera 17, and the left-eye camera 16 and the right-eye camera 17 are symmetrically distributed about the wide-angle camera 9.

The patient uses the non-contact automatic detection tonometer, the voice module 8 will play to prompt the patient to operate, and at the same time, the screen 11 will perform corresponding animation demonstration. According to the prompt, the patient places the chin on the chin rest 4 and places the forehead close to the forehead cushion 3. The weight sensor 6 detects the weight, and when the weight is lower than the pressure threshold value, the voice module 8 and the screen 11 can correspondingly prompt and demonstrate. The prompt content of the voice module 8 is "please put the chin on the chin rest 4 for detection accuracy, and keep it still". The capacitive sensor 5 will detect whether there is skin contact or not, and when there is skin contact, the capacitive sensor 5 will change from a low level to a high level. If the capacitive sensor 5 does not detect skin, the voice module 8 and the screen 11 will prompt and demonstrate accordingly. The prompting content of the voice module 8 is "please attach the forehead to the forehead cushion 3 for detection accuracy and keep the forehead motionless"; when the sensors of the mandibular rest 4 and the forehead cushion 3 detect that someone is acting, the tonometer will go to the next detection. The response time of the weight sensor 6 and the capacitance sensor 5 is in the microsecond level, and the time of the using process is not influenced. The above procedure ensures that the patient maintains a standard test posture during the test, facilitating measurement of ocular tension.

The eyeball detection flow is divided into a coarse adjustment process and a fine adjustment process. The rough adjustment process is that a wide-angle camera 9 (the selected camera has strong self-adaptation capability and can recognize human eyes under different light environments) recognizes the human face first, and then the approximate position of the human eyes is judged. After the rough position is found, the image coordinate system is converted into a physical coordinate system, and the three-axis moving platform 7 of the tonometer is moved to the vicinity of the target. At this time, the binocular camera 10 performs fine adjustment. At this time, the triaxial moving platform 7 is further aligned with the eyeball. When the images seen by the binocular camera 10 are consistent, the center of the eyeballs are aligned, the triaxial moving platform 7 stops moving, and the next procedure can be entered. The invention can greatly reduce the time of medical treatment for hospitals and patients and reduce the cost of people for hospitals.

Claims (10)

1. A method for fully automatically detecting the center of an eyeball for fixing the head of a tonometer, which is characterized by comprising the following steps:

firstly, a voice module (8) and a screen (11) on the tonometer are demonstrated to prompt a user to place a chin on a chin rest (4) and press the forehead close to a forehead cushion (3);

step two, a weight sensor (6) in the mandibular rest (4) detects weight, and corresponding prompt and demonstration are carried out through a voice module (8) and a screen (11) if necessary;

step three, a capacitive sensor (5) in the forehead cushion (3) detects whether skin contact exists, and corresponding prompt and demonstration are carried out through a voice module (8) and a screen (11) if necessary;

step four, after the head position is accurately adjusted, the wide-angle camera (9) firstly recognizes the human face, and then judges the approximate position of human eyes;

converting the image coordinate system into a physical coordinate system, and moving the triaxial moving platform (7) to the vicinity of the target;

step six, the binocular camera (10) recognizes eyeballs, the triaxial moving platform (7) performs eyeball alignment operation, when images seen by the binocular camera (10) at the moment are consistent, the centers of the eyeballs are aligned, the triaxial moving platform (7) stops moving, and the next intraocular pressure detection flow can be entered.

2. A method for the full-automatic detection of the centre of the eye for the fixation of the head of a tonometer according to claim 1, characterized in that in step two, when the weight detected by the weight sensor (6) is lower than the set pressure threshold, the voice module (8) and the screen (11) will give corresponding prompts and demonstrations.

3. A method for a tonometer head fixation full automatic detection of the centre of the eye as claimed in claim 2, wherein the pressure threshold is set to 400g.

4. A method for the full automatic detection of the centre of the eye for tonometer head fixation according to claim 1, characterized in that in step three, when the capacitive sensor (5) does not detect the skin, the voice module (8) and the screen (11) will give corresponding prompts and demonstrations.

5. The method for the head-mounted full-automatic detection of the center of an eyeball of a tonometer according to claim 1 or 2 or 3 or 4 or 5, characterized in that the tonometer comprises a detection main body (1), a head placement area (2) is arranged on the side edge of the detection main body (1), a forehead cushion (3) and a mandibular support (4) are arranged in the head placement area (2), the forehead cushion (3) is fixedly connected with the top of the head placement area (2), a capacitance sensor (5) is arranged in the forehead cushion (3), a weight sensor (6) is arranged at the bottom of the head placement area (2), the mandibular support (4) acts on the weight sensor (6) and is detachably connected with the bottom of the head placement area (2), a control main board, a triaxial mobile platform (7) and a voice module (8) are detachably connected in the detection main body (1), a wide-angle camera (9) and a binocular camera (10) which are corresponding to the head placement area (2) are arranged on the triaxial mobile platform (7), a wide-angle camera screen (9) and a binocular camera (10) are connected with the wide-angle camera screen (7), and the voice module (8) are connected with the weight sensor (7), and the weight sensor (5) are connected with the camera module (8) The binocular camera (10) and the screen (11) are electrically connected with the control main board.

6. The method for fully automatic detection of eyeball center for tonometer head fixation according to claim 5 is characterized in that a support column (12) is arranged on the detection main body (1), one end of the support column (12) is located in the detection main body (1), the other end of the support column (12) is located outside the detection main body (1), the support column (12) is connected with the detection main body (1) in an up-down sliding mode, the weight sensor (6) is detachably connected with one end of the support column (12) located outside the detection main body (1), and the mandibular support (4) is connected with one end of the support column (12) located outside the detection main body (1) and acts on the weight sensor (6).

7. The method for fully automatic detection of eyeball center for tonometer head fixation according to claim 5 is characterized in that a connecting shaft (13) is arranged on the mandibular support (4), one end of the connecting shaft (13) is fixedly connected with the center of the mandibular support (4), the other end of the connecting shaft (13) is connected with the weight sensor (6), the connecting shaft (13) is detachably connected with the inside of the supporting column (12), and the mandibular support (4) is positioned outside the supporting column (12).

8. The method for fully automatic detection of eyeball center for tonometer head fixation according to claim 6 is characterized in that a screw motor (14) and a lifting table (15) are arranged in the detection main body (1), the screw motor (14) is electrically connected with the control main board, the screw motor (14) is detachably connected with the detection main body (1), the output end of the screw motor (14) is connected with the lifting table (15), the lifting table (15) is driven by the screw motor (14) to slide up and down, and one end of the support column (12) positioned in the detection main body (1) is detachably connected with the lifting table (15).

9. A method for the head-mounted full-automatic detection of the center of the eye according to claim 5, characterized in that said binocular camera (10) is an infrared camera.

10. The method for fully automatic eye center detection for tonometer head fixation according to claim 5, wherein the wide-angle camera (9) is located above the binocular camera (10), the mandibular tray (4) is located below the binocular camera (10), the binocular camera (10) comprises a left eye camera (16) and a right eye camera (17), and the left eye camera (16) and the right eye camera (17) are distributed in bilateral symmetry with the wide-angle camera (9) as the center.