Posture detection system and method

Abstract

A posture detection system and method are provided with a detection platform device, wherein a transparent plane is formed on the top of the platform body, a light source is provided around the transparent plane, an electric slide rail with a slider on its top to set a foot lens module for upward photography is provided directly below the transparent plane, and a distance sensor for controlling the slider of the electric slide rail to follow the person is provided next to the transparent plane. When a person to be tested walks over the transparent plane, the distance sensor detects the distance of the person to be tested and controls the slider to follow the person, so that the foot lens module takes an image of the foot sole stepping on the transparent plane and transmits it to a host for real-time gait analysis, thereby generating a foot sole pressure distribution map. Finally, the analysis result can be displayed on a software interface or stored as a result analysis electronic file. During the detection process, the person to be tested does not need to wear a sensor, there will be no problem of sensor loss, and the person to be tested feels more comfortable during the detection process.

Description

Technical Field

[0001] The present invention relates to an image detection and analysis device, and in particular to a posture detection system and method. Background Art

[0002] In the existing technology, if it is necessary to measure the walking posture of the subject, such as the gait cycle, knee bending angle or plantar pressure distribution information, the subject is mainly equipped with sensors that directly measure the joint movement angle or plantar pressure, such as six-axis, nine-axis, pressure sensing and other sensors to analyze the gait and obtain information such as gait cycle, joint bending angle, plantar pressure distribution and so on.

[0003] Although the above-mentioned gait measurement method of wearing sensors on the subject can measure the subject's walking posture, since each sensor needs to be worn on the subject's joints and feet during the test, the use process will not only cause the sensor to be worn, shortening the service life of the sensor, and the sensor power consumption requires frequent charging, but also the use of sensors that need to be worn and tied to the joints during the test process may easily make the subject feel uncomfortable. Summary of the invention

[0004] Since the existing walking detection system requires the subject to wear sensors on the joints, it causes the disadvantages of shortening the service life of the sensors and making the subject uncomfortable. Therefore, the present invention uses a camera module that works with the person to cooperate with an image processing device to achieve the effect of detecting posture without wearing sensors, so that the detection process is more comfortable.

[0005] To achieve the above-mentioned object, the present invention provides a posture detection system, comprising a detection platform device and a host receiving images sensed by the detection platform device and performing real-time gait analysis, wherein:

[0006] The detection platform device is provided with a platform body, a following device, a light source and a sole lens module. A transparent plane extending in a straight line is formed on the top of the platform body. The following device is provided with an electric slide rail extending in a straight line directly below the transparent plane. A slider is provided on the top of the electric slide rail. A motor controller electrically connected to the electric slide rail is provided. A distance sensor is electrically connected to the motor controller. The distance sensor is located beside the transparent plane and in the extending direction of the transparent plane. The light source is provided around the transparent plane to illuminate the transparent platform. The sole lens module is installed on the slider and takes photos upward toward the transparent plane. The distance sensor senses the distance toward the space above the transparent plane, and drives the slider to move along the electric slide rail to follow the person through the motor controller.

[0007] The host is installed with a detection software, which receives the image taken by the sole camera module, performs real-time image processing, and generates a sole pressure distribution map.

[0008] Furthermore, the platform body of the present invention is an internally hollowed-out platform, and an extension arm is combined with the slider, the extension direction of the extension arm is perpendicular to the extension direction of the electric slide rail, and an upwardly extending straight rod is combined with the end of the extension arm, and a side lens module is combined with the portion of the straight rod higher than the transparent plane. The side lens module takes pictures inwardly toward the top of the transparent plane, and the detection software receives the image taken by the side lens module for real-time image processing to generate a knee joint angle.

[0009] Furthermore, the present invention erects a bracket on the floor next to one end of the transparent plane, and the bracket can be one or two. The distance sensor is combined with the bracket, and a short-side lens module is combined at a position where the bracket is higher than the transparent plane. The distance sensor and the short-side lens module can be combined with the same bracket or respectively combined with different brackets. The short-side lens module takes a picture along the extension direction of the transparent plane toward the top of the transparent plane. The detection software receives the image taken by the short-side lens module and performs real-time image processing to generate a shoulder tilt angle, and can further calculate information such as body tilt and head shaking.

[0010] Preferably, the present invention displays the image processed by the detection software, the plantar pressure distribution diagram, the knee joint angle and the shoulder tilt angle on a software interface, and selectively stores the image processed by the detection software, the plantar pressure distribution diagram, the knee joint angle and the shoulder tilt angle on the host as a result analysis electronic file.

[0011] When the present invention is used, a subject is allowed to walk along a straight line across the transparent plane on the top of the platform body. During the process of the subject walking across the transparent plane, the distance measuring sensor will sense the distance of the subject, so that the motor controller drives the slider to move along the electric slide rail with the person, so that the sole camera module can capture the image of the sole of the subject stepping on the transparent plane over time, and the detection software receives the image captured by the sole camera module for real-time image processing to generate a sole pressure distribution map; the side camera module further provided in the present invention can capture the side image of the subject during the process of the subject walking across the transparent plane, so that the detection software receives the image captured by the side camera module for real-time image processing to generate the knee joint angle.

[0012] The effectiveness of the present invention lies in that, since the sole lens module of the present invention moves with the person (the side lens module further provided also moves with the person), it can continuously capture images of the soles of the subject's feet stepping on the transparent plane when walking for the purpose of image processing, and during the detection process, the subject does not need to wear sensors, there will be no problem of sensor loss, and the subject feels more comfortable during the detection process. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 is a stereogram of a preferred embodiment of the present invention;

[0014] Figure 2 is a front view of a preferred embodiment of the present invention;

[0015] Figure 3 is a top view of a preferred embodiment of the present invention;

[0016] Figure 4 is a schematic diagram of a software interface of a preferred embodiment of the present invention;

[0017] Figure 5 is a schematic diagram of an electronic file of result analysis of a preferred embodiment of the present invention;

[0018] Figure 6 It is a schematic diagram of the operation of using and implementing a preferred embodiment of the present invention;

[0019] Figure 7 is a flow chart of the steps of processing plantar images in a preferred embodiment of the present invention;

[0020] Figure 8 is a flow chart of the steps of processing the silhouette image in a preferred embodiment of the present invention;

[0021] Fig. 9 It is a flow chart of the steps of processing front / back images in a preferred embodiment of the present invention.

[0022] Explanation of symbols:

[0023] 10 detection platform device 11 platform body

[0024] 111 transparent plane 112 long perforation

[0025] 113 stairs 114 railing

[0026] 12 people installation 121 electric slide

[0027] 1211 Slider 122 Motor Controller

[0028] 123 bracket 124 distance sensor

[0029] 13 light source 14 foot lens module

[0030] 141 lens module 15 side lens module

[0031] 151 extension arm 152 straight rod

[0032] 16 short side lens modules 20 host

[0033] 21 Detection software 23 Software interface

[0034] 24 Result analysis electronic file 30 Foot pressure distribution diagram

[0035] 31 knee joint angle 32 shoulder tilt angle

[0036] 40 Subject A Ground

[0037] B marker DETAILED DESCRIPTION

[0038] In order to understand the technical features and practical effects of the present invention in detail and to implement it according to the contents of the specification, a preferred embodiment as shown in the drawings is further described in detail as follows.

[0039] Please see Figures 1 to 4 The preferred embodiment of the present invention shown in FIG. 1 provides a posture detection system for performing posture detection on a subject, comprising a detection platform device 10 disposed on a ground A when in use, and a host 20 for receiving images sensed by the detection platform device 10 and performing real-time gait analysis, wherein:

[0040] The detection platform device 10 is provided with a platform body 11, a follower device 12, a light source 13, a sole lens module 14, a side lens module 15 and a short side lens module 16. The platform body 11 is a rectangular platform with a hollow interior and is arranged on the ground A. A transparent plane 111 extending left and right in a straight line direction is formed on the rear side of the top of the platform body 11. The length of the transparent platform 111 can be established in a range between 50 cm and 3 meters. A step 113 is provided on the left edge of the platform body 11, and a railing 114 is provided on one side of the top surface of the platform body 11.

[0041] The following device 12 is provided with an electric slide rail 121 extending left and right in a straight direction directly below the transparent plane 111. The electric slide rail 121 is set on the ground A. A slider 1211 is provided on the top of the electric slide rail 121 in a manner that can be controlled to slide along the electric slide rail 121. A motor controller 122 is provided in conjunction with the electric slide rail 121. The motor controller 122 is electrically connected to the electric slide rail 121. The motor controller 122 can be selectively set inside or outside the platform body 11. For example, in the preferred embodiment, the motor controller 122 is set on the ground A outside the platform body 11 and is located next to the right end of the transparent plane 111.

[0042] A bracket 123 is provided on the outside of the platform body 11. The bracket 123 is erected on the short side of the transparent plane 111 and can be set up separately or combined with the motor controller 122; a distance sensor 124 is combined with the bracket 123. The distance sensor 124 is located in the extension direction of the transparent plane 111 and senses the distance toward the space above the transparent plane 111, that is, the space where the subject will walk. The distance data is given to the electrically connected motor controller 122, or the distance data is given to the electrically connected host 20 to control the motor controller 122, so that the motor controller 122 drives the slider 1211 to move along the electric slide rail 121 with the person.

[0043] The light source 13 is arranged around the transparent plane 111 to illuminate the transparent plane 111. For example, in the preferred embodiment of the present invention, the light source 13 is combined with a light strip at the edge of the transparent plane 111 to illuminate the transparent plane 111. The sole lens module 14 is installed on the slider 1211 to shoot upward in the direction of the transparent plane 111. The sole lens module 14 may include more than one lens module 141. For example, in the preferred embodiment of the present invention, the sole lens module 14 includes two lens modules 141 spaced apart along the extension direction of the short sides of the transparent plane 111.

[0044] The side lens module 15 is coupled to an extension arm 151 on the slider, wherein the extension direction of the extension arm 151 is perpendicular to the extension direction of the electric slide rail 121, and an upwardly extending straight rod 152 is coupled to the end of the extension arm 151. The side lens module 15 is coupled to a portion of the straight rod 152 that is higher than the transparent plane 111. The side lens module 15 shoots inward at a horizontal shooting angle toward the top of the transparent plane 111, and the height of the side lens module 15 is approximately the height of the subject's knees.

[0045] The short-side lens module 16 is combined with the bracket 123 , and the distance sensor 124 is located at a height higher than the short-side lens module 16 . The short-side lens module 16 takes photos along the extending direction of the transparent plane 111 toward the top of the transparent plane 111 .

[0046] The host 20 is internally installed with a detection software 21 for executing the posture detection method. The detection software 21 receives the images taken by the sole lens module 14, the side lens module 15 and the short side lens module 16 and performs real-time image processing. For the images taken by the sole lens module 14, the side lens module 15 and the short side lens module 16, sole image processing, side image processing and front / back image processing are performed. Figure 4 As shown, a plantar pressure distribution diagram 30, a knee joint angle 31 and a shoulder tilt angle 32 are generated, and the plantar pressure distribution diagram 30, the knee joint angle 31 and the shoulder tilt angle 32 data are displayed together with the processed image on a software interface 23; and the image processed by the detection software 21, that is, the image taken by the plantar lens module 14, the side lens module 15 and the short side lens module 16, and the plantar pressure distribution diagram, the knee joint angle and the shoulder tilt angle data can be selectively displayed, as shown in FIG. Figure 5 As shown, the result analysis electronic file 24 is stored in the host 20 .

[0047] Please see Figure 1 and Figure 6 As shown, when the subject 40 is performing posture detection in the posture detection system of the preferred embodiment of the present invention, a marker B is firstly set on the shoulders, waist, knee joints and ankles of the subject 40 by gluing or wearing. For example, in the preferred embodiment, each marker B is a pink and an orange cloth. The color of the marker B can be changed according to the color of the clothes of the subject 40. Then, the subject 40 is allowed to walk along a straight line through the transparent plane 111 on the top of the platform body 11. The subject 40 walks through the transparent plane During the process of 111, the distance sensor 124 senses the distance of the subject 40, so that the motor controller 122 drives the slider 1211 to move along the electric slide rail 121 with the person, so that the sole camera module 14 captures the sole image of the subject 40 stepping on the transparent plane 111 over time, the side camera module 15 captures the side image of the subject 40 over time, and the short side camera module 16 captures the front / back image of the subject 40 over time.

[0048] When the detection software 21 processes the sole image captured by the sole camera module 14, Figure 7As shown in the flowchart of the steps, an image is first input, and fisheye image correction is performed. The plantar image is captured using the difference between the green grayscale value and the red grayscale value and the blue grayscale value of each pixel. Then, a median filter is used to filter out noise that is not in the plantar image, and the plantar image is converted into a grayscale image. The maximum grayscale value is found, and the plantar pressure distribution grayscale value is set according to the interval of each pixel grayscale value relative to the maximum grayscale value. Finally, the image is output as the plantar pressure distribution map 30.

[0049] When the detection software 21 processes the side image captured by the side camera module 15, Figure 8 As shown in the flowchart of the steps, the image is first input, fisheye image correction is performed, three markers at the waist, knee joint, and ankle are identified, and the individual center points of the three markers are calculated. Then the three center points are connected and the knee joint angle is calculated. Finally, the image is output as a real-time screen as the knee joint angle 31, and the analysis diagram file and data are stored.

[0050] When the detection software 21 processes the front / back images captured by the side camera module 15, Fig. 9 As shown in the flowchart of the steps, an image is first input, the markers on the two shoulders are identified, and the individual center points of the two shoulder markers are calculated. Then, the angle between the two points and the horizontal line is calculated to calculate the shoulder tilt angle. Finally, the image is output as a real-time screen as the shoulder tilt angle 32, and the analysis chart file and data are stored.

[0051] The detection software 21 can further calculate and analyze gait (including step length, step width, step frequency, proportion of each stage of the support phase and swing phase in the walking cycle, left-right symmetry, some or all items), plantar pressure center trajectory, body tilt, head shaking and other information, and store them in the host 20 as a result analysis electronic file 24.

[0052] The above description is only a preferred embodiment of the present invention and is not intended to limit the scope of rights claimed by the present invention. Any other equivalent changes or modifications that do not deviate from the spirit disclosed by the present invention should be included in the scope of the patent application of the present invention.

Claims

1. A posture detection system, characterized in that: The invention comprises a detection platform device and a host receiving images sensed by the detection platform device to perform real-time gait analysis, wherein: The detection platform device is provided with a platform body, a follower device, a light source and a sole lens module. A transparent plane extending in a straight line is formed on the top of the platform body. The follower device is provided with an electric slide rail extending in a straight line directly below the transparent plane. A slider is provided on the top of the electric slide rail. A motor controller electrically connected to the electric slide rail is provided. A distance sensor is electrically connected to the motor controller. The distance sensor is located beside the transparent plane and in the extension direction of the transparent plane. The distance sensor is provided on a bracket at one end of the transparent plane. The bracket is erected on the transparent plane. The bracket is provided with a short side lens module at a position higher than the transparent plane, the short side lens module takes photos along the extension direction of the transparent plane toward the top of the transparent plane, the light source is arranged around the transparent plane to illuminate the transparent plane, the sole lens module is installed on the slider and takes photos upward toward the transparent plane, the slider is provided with an extension arm, the extension direction of the extension arm is perpendicular to the extension direction of the electric slide rail, a straight rod extending upward is provided at the end of the extension arm, a side lens module is provided at a portion of the straight rod higher than the transparent plane, and the side lens module takes photos inward toward the top of the transparent plane; wherein: The distance measuring sensor senses the distance of the subject toward the space above the transparent plane and provides distance data to a host computer that is electrically connected; After receiving the distance data, the host controls the motor controller to drive the slider to move along the electric slide rail to move with the person; The host is installed with a detection software, which synchronously receives the images taken by the sole lens module, the side lens module and the short side lens module for real-time image processing to generate a sole pressure distribution diagram, a knee joint angle and a shoulder tilt angle.

2. The posture detection system according to claim 1, characterized in that: The image processed by the detection software, the plantar pressure distribution diagram, the knee joint angle and the shoulder tilt angle are displayed on a software interface.

3. The posture detection system as claimed in claim 2, characterized in that: The image processed by the detection software, the plantar pressure distribution diagram, the knee joint angle and the shoulder tilt angle are stored in the host as an electronic file for result analysis.

4. The posture detection system as claimed in claim 3, characterized in that: The detection software can calculate and analyze information such as images, gait, plantar pressure center trajectory, body tilt, head shaking, etc., and store the information such as images, gait, plantar pressure center trajectory, body tilt, head shaking, etc. processed by the detection software in the host as an electronic file of result analysis.

5. The posture detection system according to claim 1, characterized in that: The length of the transparent plane is in the range of 50 centimeters to 3 meters.

6. A posture detection method, characterized in that: The method is implemented in the posture detection system as claimed in claim 1 and the steps include: Input image; Perform fisheye image correction; The foot sole image is captured by using the difference between the green grayscale value and the red grayscale value and the blue grayscale value of each pixel; Use a median filter to filter out noise in non-plantar images; Convert the plantar image into grayscale and find the maximum grayscale value; According to the interval of each pixel grayscale value relative to the maximum grayscale value, the grayscale value of the plantar pressure distribution is set; and The image is output as the plantar pressure distribution map.

7. A posture detection method, characterized in that: The method is implemented in the posture detection system as claimed in claim 1 and the steps include: Input image; Perform fisheye image correction; Identify the three markers at the waist, knee joint, and ankle, and calculate the individual center points of the three markers; Connect the three center points and calculate the knee joint angle; The image is output in real time as the knee joint angle, and the analysis chart file and data are stored.

8. A posture detection method, characterized in that: The method is implemented in the posture detection system as claimed in claim 1 and the steps include: Input image; Identify the markers on both shoulders and calculate the center points of the markers on both shoulders; Calculate the angle between the two points and the horizontal line to calculate the shoulder tilt angle; and The image is output in real time as the shoulder tilt angle, and the analysis chart file and data are stored.

Images