WO2017193590A1 - 仰卧起坐测试的计数方法及*** - Google Patents

仰卧起坐测试的计数方法及*** Download PDF

Info

Publication number
WO2017193590A1
WO2017193590A1 PCT/CN2016/112276 CN2016112276W WO2017193590A1 WO 2017193590 A1 WO2017193590 A1 WO 2017193590A1 CN 2016112276 W CN2016112276 W CN 2016112276W WO 2017193590 A1 WO2017193590 A1 WO 2017193590A1
Authority
WO
WIPO (PCT)
Prior art keywords
test
area
tester
texture
frame
Prior art date
Application number
PCT/CN2016/112276
Other languages
English (en)
French (fr)
Inventor
刘远民
柳庆红
许宏强
Original Assignee
深圳泰山体育科技股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳泰山体育科技股份有限公司 filed Critical 深圳泰山体育科技股份有限公司
Publication of WO2017193590A1 publication Critical patent/WO2017193590A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/20—Movements or behaviour, e.g. gesture recognition
    • G06V40/23—Recognition of whole body movements, e.g. for sport training
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/20—Movements or behaviour, e.g. gesture recognition
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/40—Analysis of texture

Definitions

  • the present invention relates to the field of image recognition processing technology, and more particularly to an application of a sit-up test count.
  • the counting method is usually manual counting, but the testing efficiency is not high, and the error rate is large, and the repeated operation for a long time is also likely to cause the tester to be exhausted, and the irregular operation is not easy to detect. It is now less used for body testing. On the other hand, the start of the 1 minute test requires additional personnel, such as pressing the chronograph button or shouting the password, the degree of automation is not high.
  • the present invention provides a counting method for sit-ups, including:
  • an identification area having a specific texture is set in the test area, and the identification area is on the tester at the time of testing The area where the half body is located, and the camera is set obliquely above the identification area;
  • the texture image obtained by the analysis is set to the start frame when the pixel value of the texture in the image of the frame reaches the preset sleep position threshold for the first time;
  • step S3 further includes: starting the timing of setting the motion start frame; and step S4, before extracting the texture image of the frame, determining whether the time relative to the timing of the frame has reached the preset time Set the duration, then end the test, otherwise continue testing.
  • the specific texture disposed in the identification area is at least one linear identification pattern, and the extending direction of the linear identification pattern is substantially perpendicular to the supine direction of the tester, and the pixel value of the texture in the texture image analyzed in each step is The length value of the line recognition pattern.
  • the step S1 further includes: setting a foot positioning area for the tester's foot to be placed in the test area, and a leg placement area between the foot positioning area and the identification area, and the leg placement area is spaced apart
  • the two positioning lines extending substantially along the tester's supine direction, the tester's legs are placed along the two positioning lines during the test; the tester's height data is obtained before the test, and the tester is determined to make a sit-up according to the height data.
  • a line-shaped identification pattern in which the recognition area changes during operation is determined as an associated line pattern; when the lines in the texture image are analyzed in steps S3 and S4, only the associated line pattern is analyzed.
  • the number of the associated line patterns is plural, and the average values of the lengths of all associated line patterns are analyzed in steps S3 and S4, and the lying position threshold is all that the camera can capture when the tester's shoulder blade contacts the test area.
  • the average value of the length of the associated line pattern, the sitting posture is the time when the tester is completely seated
  • the invention also provides a counting system for sit-ups, comprising:
  • test area comprising an identification area having a specific texture, the identification area being an area of the upper body of the tester at the time of testing;
  • a camera which is disposed obliquely above the identification area, continuously captures a test image including the tester and the recognition area at a certain frame rate, and obtains a texture image of the identification area;
  • a start frame acquisition module which extracts and analyzes a texture image obtained by the camera, and sets the frame as a start frame when the pixel value of the texture in the frame texture image reaches a preset sleep position threshold for the first time;
  • the texture image analysis module further extracts the texture image of each frame subsequent to the analysis start frame, and sends a counting instruction whenever the pixel value of the texture in the texture image is gradually increased from the preset lying posture threshold to the preset sitting posture threshold. ;
  • the counting module counts the number of sit-ups once after receiving the counting instruction issued by the texture image analyzing module.
  • the start frame acquisition module further issues a start command while setting the start frame
  • the counting system further includes:
  • timing module which starts timing when receiving a start instruction issued by the start frame acquisition module, and subsequently records in real time the duration data of the current frame relative timing start time;
  • the test end determination module which retrieves the duration data recorded by the timing module, determines whether the duration data corresponding to the current frame has reached the preset duration, and then issues an end instruction, otherwise issues a resume instruction; the texture image analysis module extracts and analyzes a frame. Before the texture image, the instruction sent by the test end determination module is received to end the test according to the end instruction, or continue the test according to the resume instruction.
  • the specific texture set in the identification area is at least one line identification pattern, and the line The extending direction of the pattern recognition pattern is substantially perpendicular to the tester's supine direction, and the pixel value of the texture in the texture image analyzed by the start frame acquisition module and the texture image analysis module is the length value of the line recognition pattern.
  • a foot positioning area for the tester's foot is set in the test area, and a leg placement area is disposed between the foot positioning area and the identification area, and the leg placement area is spaced substantially along the tester.
  • Two positioning lines extending in the supine direction, the tester's legs are respectively placed along two positioning lines during the test;
  • the counting system further includes a data correlation module for obtaining the height data of the tester before the test, and determining the test according to the height data.
  • the linear recognition pattern in which the recognition area changes when the sit-up action is performed is defined as an associated line pattern; when the start frame acquisition module and the texture image analysis module analyze the texture in the texture image, only the associated line pattern is analyzed.
  • the number of associated line patterns is multiple, and the start frame acquisition module and the texture image analysis module analyze the average of the lengths of all associated line patterns, and the position threshold is the camera can take when the tester's shoulder blade contacts the test area.
  • the average of the lengths of all associated line patterns arriving, the sitting threshold is the average of the lengths of all associated line patterns that the camera can capture when the tester is fully seated.
  • self-service sit-up test and counting can be realized by image recognition processing technology.
  • the sit-up test process can be automated, reducing manual intervention, significantly improving test efficiency, and a good test experience.
  • FIG. 1 is a schematic flow chart of a sit-up counting method provided by the present invention
  • FIG. 2 is a schematic view of a test area in an embodiment of the present invention.
  • FIG. 3 is a schematic diagram of a sit-up counting system provided by the present invention.
  • the implementation and application of the present invention is primarily, but not exclusively, limited to the National Physique Measurement Program.
  • sit-ups reflect the strength and continuous working ability of the human waist and abdomen muscles.
  • the tester is required to lie on the horizontal surface, and the legs are slightly separated from the knees by 90°. The fingers are folded over the back of the head, and the feet are pressed to fix the legs.
  • the tester quickly lifted up, the elbow touched or exceeded the knees, and then returned to the supine position, the shoulders hit the horizontal plane to complete once, and so on, and the number of consecutive sit-ups was accumulated. It should be noted that if the tester borrows the strength of the elbow support to complete the sit-up or the elbow does not touch the knees, the situation is not counted. In general, the test time for sit-ups is 1 minute.
  • the present invention achieves automatic recognition and counting of sit-ups by means of image recognition processing.
  • the present invention presupposes that the tester performs the action according to the specification.
  • the present invention provides a flow of a sit-up counting method as shown in FIG.
  • step S1 an identification area having a specific texture is set in the test area, and the identification area is an area of the upper body of the tester at the time of the test, and a camera is disposed obliquely above the identification area.
  • Test area 210 includes, but is not limited to, providing a sit-up test for the tester.
  • the identification area 211, the foot positioning area 212 and the leg placement area 213 are arranged, as shown by the black dotted line in the figure, respectively, to indicate that the upper body, the foot and the legs should be placed during the tester test.
  • the leg placement area 213 is disposed between the foot positioning area 212 and the identification area 211.
  • the leg placement area 213 is not limited to be in contact with the identification area 211, and the leg placement area 213 is narrower than the identification area 211 in the direction in which the width W is extended, in order to distinguish the positions of the legs from the upper body, indicating the supine direction.
  • the two sides of the leg placement area 213 are spaced apart from each other with a positioning line 2131a and a positioning line 2131b extending substantially in the supine direction of the tester.
  • the legs of the examiner are placed along the positioning line 2131a and the positioning line 2131b, respectively.
  • the foot positioning area 212 is provided with a fixing device for pressing the feet to fix the legs during the test.
  • the identification area 211 further includes at least one specific texture 2111 extending in a direction substantially perpendicular to the tester's supine direction.
  • the pattern of the specific texture of the present invention is not limited.
  • the specific texture may also have a linear pattern of uniformly distributed curves, broken lines, arcs, etc. It may be a pattern consisting of regular geometric figures, or a combination of light and dark geometric figures, and the like.
  • the specific texture 2111 inside the identification area 211 is a plurality of linear identification patterns in a straight line as shown in FIG. 2, preferably, the height data of the tester is obtained before the test, and the tester makes the determination according to the height data.
  • a linear recognition pattern that changes in the recognition area 211 that is, a specific texture that reflects the upper body movement of the tester, is defined as an associated line pattern.
  • testers of different heights may block multiple specific lines in different ranges when lying on their backs. When the tester is from supine to sitting, these specific lines will gradually reappear in the captured image; when the tester is restored to the supine position by sitting again These specific lines will gradually be blocked by the upper body.
  • the dynamic change of the associated line pattern objectively reflects the tester's motion process.
  • the selection of the associated line pattern can not only ensure that the tester must reach the judgment threshold required for supine and sitting during the test, and prevent the cheating phenomenon during self-test, and also reduce the data processing. the amount.
  • a camera (not shown in FIG. 2) is provided for image capturing of the recognition area 211.
  • the position of the camera is preferably placed on a side substantially perpendicular to the tester's supine direction, as illustrated by the field of view of FIG.
  • step S2 the test image including the tester and the recognition area is continuously captured by the camera at a certain frame rate, and the texture image in the identification area is obtained.
  • the tester prepares for the sit-up test in the test area 210 shown in FIG. 2, the foregoing may be used.
  • the camera continues to take images for the tester. Since the shooting of the camera has a specific frame rate, a number of test images including the tester and the recognition area can be obtained in time series. In the subsequent image processing step, the specific texture in the identification area will be analyzed. In the case where the aforementioned associated line pattern is determined, only the associated line pattern may be analyzed, that is, in the sit-up process based on the associated line pattern. The length change of the captured image identifies the number of sit-ups completed by the tester, which will be detailed in later steps.
  • step S3 the texture image obtained by the analysis is set to the start frame when the pixel value of the texture in the frame texture image reaches the preset sleep position threshold for the first time.
  • the accuracy of the test method of the present invention can be improved by effectively determining the starting frame of the sit-up test. Therefore, in an optional embodiment, the associated line pattern in the test image is analyzed in real time from the time when the camera starts to acquire the test image frame by frame, specifically, the average value of the length of the plurality of associated line patterns is analyzed. That is to say, in the process of sitting or lying down, the upper body will block the associated line pattern to different degrees, and in the image acquired by the camera, the lengths of the plurality of associated line patterns are increased or decreased to different degrees. Small, for the overall analysis, the average value of the plurality of linear lengths included is analyzed.
  • a one-position threshold is first preset as a judgment point of the timing start frame.
  • the prone position threshold is the average of the lengths of all associated line patterns that the camera can capture when the tester's shoulder blade contacts the test area. Since the length is specifically represented by a plurality of pixels in the test image, the length of the line pattern referred to in the present invention can be equivalently interpreted as a line pattern pixel value, and the average length of the associated line pattern can also be referred to as a pixel average. value.
  • the average length of the associated line pattern reaches the preset lying position threshold for the first time, the tester is deemed to be ready for testing, and the current frame is set as the start frame, and the timing starts.
  • step S4 the texture image of each frame subsequent to the initial frame is continuously extracted, and the pixel value of the texture in the texture image is counted once when the pixel value of the texture is gradually increased from the preset lying posture threshold to the preset sitting posture threshold.
  • the present invention presets a sitting posture threshold, which is regarded as a judgment point for the tester to sit.
  • the sitting threshold is the average of the lengths of all associated line patterns that the camera can capture when the tester is fully seated.
  • the average length of the associated line pattern in the test image gradually becomes larger until it reaches the sitting threshold.
  • the present invention enables automatic activation, recognition, and counting for a supine to sit-up.
  • the accumulation of the number of times is a repeated reproduction of the relevant steps of the method, that is, after the start frame, based on the analysis of the average value of the associated line pattern, each sitting is completed. , the number of counts is superimposed once.
  • the present invention can also provide the following methods:
  • the average time required for a sit-up test is 1 minute. While setting the motion start frame, the timing starts, and the subsequent continuous determination determines whether the current frame relative timing start time (start frame) has reached the preset duration (as in the embodiment, 1 minute), and if so, ends the test. , that is, stop extracting the current frame test image, and the time is over, and the accumulated number of sit-ups is automatically saved; if not, the test is continued.
  • the present invention also provides a sit-up counting system using the foregoing method, as shown in the system diagram of FIG. 3, comprising:
  • Test area 210 This includes, but is not limited to, providing a sit-up test for the tester.
  • the test area 210 is provided with an identification area 211 having a specific texture, a foot positioning area 212, and a leg placement area 213, respectively indicating the area where the upper body, the foot and the legs of the tester should be placed, and the foot positioning area 212 is also Includes a fixture to help the tester secure the legs during the test.
  • the layout of the three in the test area 210 is the same as that shown in FIG. 2, and details are not described herein again.
  • the camera 220 its position can be referred to as shown in the field of view of FIG. 2, which is disposed obliquely above the identification area 211 for image capturing of the identification area 211, and its position is preferably set on a side substantially perpendicular to the tester's supine direction.
  • the camera 220 continuously captures the test image including the tester and the recognition area 211 at a certain frame rate to obtain a texture image of the recognition area.
  • Data association module 230 Through the module, the counting system first directly determines the specific texture associated with the tester in the identification area 211. Specifically, the data association module 230 first determines, according to the tester's height data, a plurality of linear recognition patterns in which the length value changes in the recognition area 211 in the image acquired by the camera 220 when the tester performs the sit-up action, and It is defined as an associated line pattern. The reason for selecting the associated line pattern can be referred to the related description in the foregoing description of the counting method, and thus will not be described herein. Subsequent modules of the system will analyze the changes in the average length value of the associated line pattern captured during the sit-up process, objectively reflecting the tester's motion process. The setting of data association module 230 is not intended to limit the counting system of the present invention.
  • the start frame acquisition module 240 extracts and analyzes the texture image obtained by the camera.
  • the frame is set as the start frame.
  • a one-position threshold is preset as a judgment point of the timing start frame.
  • the prone position threshold is the average of the lengths of all associated line patterns that the camera can capture when the tester's shoulder blade contacts the test area.
  • the average length of the associated line pattern reaches the preset lying position threshold, it is regarded as the tester.
  • Prepare for the test set the current frame as the start frame, and generate the connection instruction. You can also issue a start command at the same time.
  • the texture image analysis module 250 continues to extract the texture image of each frame subsequent to the analysis start frame, and sends a counting instruction whenever the pixel value of the texture in the texture image is gradually increased from the preset lying posture threshold to the preset sitting posture threshold. .
  • a sitting posture threshold is preset, which is regarded as a judgment point of the tester sitting.
  • the sitting threshold is the average of the lengths of all associated line patterns that the camera can capture when the tester is fully seated.
  • the texture image analysis module 250 receives the connection instruction generated by the start frame acquisition module 240, and continues to extract the texture image of the subsequent frames obtained by the analysis camera until it is found that the average value of all associated line patterns in the texture image is preset by the prone position. When the threshold is gradually increased to the preset sitting posture threshold, a counting instruction is generated.
  • the counting module 260 after receiving the counting instruction generated by the texture image analyzing module 250, counts the number of sit-ups once.
  • the system enables automatic identification and counting of a sit-up action.
  • the texture image analysis module 250 continuously analyzes the change in the length of the associated line pattern to determine whether or not to count.
  • system for the automatic start and end of the test may further include:
  • the timing module 270 when it receives the start instruction issued by the foregoing start frame acquisition module 240, starts timing, and then records the duration data of the current frame relative to the start time of the current frame in real time.
  • the test end determination module 280 the time length data recorded by the timing module 270 is retrieved, and it is determined whether the time length data corresponding to the current frame has reached a preset duration (may be 1 minute in this embodiment), and if yes, the end is issued. An instruction is issued; otherwise, the texture image analysis module 250 receives the instruction issued by the test end determination module 280 before extracting and analyzing the texture image of a certain frame, End the test according to the end command, automatically save the accumulated number of sit-ups, or continue the test according to the continuation command.
  • a preset duration may be 1 minute in this embodiment
  • the sit-up self-service test implemented by the present invention can reduce labor costs, improve test efficiency, and give testers a good test experience.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Human Computer Interaction (AREA)
  • Health & Medical Sciences (AREA)
  • Social Psychology (AREA)
  • Psychiatry (AREA)
  • Multimedia (AREA)
  • General Health & Medical Sciences (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
  • Image Analysis (AREA)

Abstract

一种仰卧起坐的计数方法及对应的计数***，该计数方法包括：S1、在测试区域设置具有特定纹路的识别区，识别区为测试时测试者的上半身所在区域，在识别区的斜上方设置摄像机；S2、利用摄像机以一定帧率持续拍摄包括测试者和识别区在内的测试图像，得到识别区内的纹路图像；S3、分析得到的纹路图像，当初次发现某帧纹路图像中纹路的像素值达到预设的卧姿阈值时，将该帧设置为起始帧；S4、继续提取分析后续各帧的纹路图像，每当发现纹路图像中纹路的像素值由预设的卧姿阈值逐渐增大至预设的坐姿阈值时计数一次。该方案可实现自助式的仰卧起坐测试与计数，测试过程无需人工干涉，提高了测试效率。

Description

仰卧起坐测试的计数方法及*** 技术领域
本发明涉及图像识别处理技术领域，尤其涉及仰卧起坐测试计数的应用。
背景技术
目前，随着社会经济发展、人民对健康认识的提高、以及国际形势的变化，国家对整体国民体质的增强已经提升到战略性的高度。在这个大背景下，迫切需要对全民体质进行更高效、准确的监测。
在国家体育总局群体司于2003年发布的《国民体质测定标准手册及标准(成年人部分)》中，仰卧起坐被列入国民体质的测试项目之一。
但是，传统的仰卧起坐实施计数方法中，通常为人工计数，但存在测试效率不高，且误差率大，长时间的反复性操作也容易造成测试员疲惫，对不规范动作不易察觉等缺点，现已较少应用于体质检测。另一方面，启动1分钟的测试需要人员额外操作，如按动计时按钮或是喊口令，自动化程度不高。
还有一种现有技术，就是通过在测试者身上佩戴电子设备来判断测试者动作是否规范，但测试者有束缚感，舒适程度低，测试体验差。
因此，面对当代人们生活质量的提高，对体质检测的越发关注，对检测中的项目测试精度要求越来越高的情形下，上述提及的仰卧起坐已有的测试方法尚不具备良好测试体验，测试效率仍有待提高。
发明内容
为解决上述技术问题，本发明提供一种仰卧起坐的计数方法，包括：
S1、在测试区域设置具有特定纹路的识别区，识别区为测试时测试者的上 半身所在区域，在识别区的斜上方设置摄像机；
S2、利用摄像机以一定帧率持续拍摄包括测试者和识别区在内的测试图像，得到识别区内的纹路图像；
S3、分析得到的纹路图像，当初次发现某帧纹路图像中纹路的像素值达到预设的卧姿阈值时，将该帧设置为起始帧；
S4、继续提取分析初始帧后续各帧的纹路图像，每当发现纹路图像中纹路的像素值由预设的卧姿阈值逐渐增大至预设的坐姿阈值时计数一次。
在可选实施例中，步骤S3还包括，在设置动作起始帧的同时，计时开始；步骤S4中，提取某帧的纹路图像之前，先判定该帧所在时刻相对计时开始时刻是否已达到预设时长，是则结束本次测试，否则继续测试。
在可选实施例中，识别区内设置的特定纹路为至少一条线状识别图案，线状识别图案的延伸方向与测试者仰卧方向基本垂直，各步骤中分析的纹路图像中纹路的像素值为线状识别图案的长度值。
进一步，步骤S1还包括：在测试区域内设定有供测试者脚部放置的脚部定位区，脚部定位区与识别区之间设置有腿部放置区，腿部放置区内间隔设置有基本沿着测试者仰卧方向延伸的两定位线条，测试时测试者的双腿分别沿着两定位线条放置；测试前获取测试者的身高数据，根据该身高数据确定出测试者做出仰卧起坐动作时识别区会发生变化的线状识别图案，定为关联线状图案；步骤S3和S4中分析纹路图像中纹路时，仅分析所述关联线状图案。
进一步，所述关联线状图案的数量为多条，步骤S3和S4中分析的是所有关联线状图案长度的平均值，卧姿阈值为测试者肩胛骨接触测试区域时摄像机所能拍到的所有关联线状图案长度的平均值，坐姿阈值为测试者完全坐起时摄 像机所能拍到的所有关联线状图案长度的平均值。
本发明还提供一种仰卧起坐的计数***，包括：
测试区域，其包括具有特定纹路的识别区，识别区为测试时测试者的上半身所在区域；
摄像机，其设置在识别区的斜上方，以一定帧率持续拍摄包括测试者和识别区在内的测试图像，得到识别区的纹路图像；
起始帧获取模块，其提取并分析摄像机得到的纹路图像，当初次发现某帧纹路图像中纹路的像素值达到预设的卧姿阈值时，将该帧设置为起始帧；
纹路图像分析模块，其继续提取分析起始帧后续各帧的纹路图像，每当发现纹路图像中纹路的像素值由预设的卧姿阈值逐渐增大至预设的坐姿阈值时，发出计数指令；
计数模块，在接收到纹路图像分析模块发出的计数指令后，对仰卧起坐次数计数一次。
在可选实施例中，起始帧获取模块在设置起始帧的同时还发出一起始指令，该计数***进一步包括：
计时模块，其在接收到起始帧获取模块发出的起始指令时开始计时，后续实时记录当前帧相对计时开始时刻的时长数据；
测试结束判定模块，其调取计时模块记录的时长数据，判定当前帧所对应的时长数据是否已达到预设时长，是则发出结束指令，否则发出继续指令；纹路图像分析模块在提取分析某帧的纹路图像之前，接收测试结束判定模块发出的指令，以根据结束指令结束本次测试，或者根据继续指令继续进行测试。
在可选实施例中，识别区内设置的特定纹路为至少一条线状识别图案，线 状识别图案的延伸方向与测试者仰卧方向基本垂直，起始帧获取模块以及纹路图像分析模块分析的纹路图像中纹路的像素值为线状识别图案的长度值。
进一步，测试区域内还设定有供测试者脚部放置的脚部定位区，脚部定位区与识别区之间设置有腿部放置区，腿部放置区内间隔设置有基本沿着测试者仰卧方向延伸的两定位线条，测试时测试者的双腿分别沿着两定位线条放置；该计数***还包括数据关联模块，用于测试前获取测试者的身高数据，根据该身高数据确定出测试者做出仰卧起坐动作时识别区会发生变化的线状识别图案，定为关联线状图案；起始帧获取模块以及纹路图像分析模块分析纹路图像中纹路时，仅分析关联线状图案。
进一步，关联线状图案的数量为多条，起始帧获取模块以及纹路图像分析模块分析的是所有关联线状图案长度的平均值，卧姿阈值为测试者肩胛骨接触测试区域时摄像机所能拍到的所有关联线状图案长度的平均值，坐姿阈值为测试者完全坐起时摄像机所能拍到的所有关联线状图案长度的平均值。
实施本发明，通过图像识别处理技术能够实现自助式的仰卧起坐测试与计数。仰卧起坐测试过程由此可实现自动化，减少人工干预，测试效率明显提高，测试者体验感好。
附图说明
图1是本发明提供的一种仰卧起坐计数方法的流程示意图；
图2是本发明实施例中测试区域示意图；
图3是本发明提供的一种仰卧起坐计数***示意图。
具体实施方式
本发明的实施和应用主要但不限于是在国民体质测定项目中。
在国民体质测定中，仰卧起坐反映人体腰腹部肌肉的力量及持续工作能力。测试前，要求测试者仰卧于水平面上，双腿稍分开屈膝呈90°，双手手指交叉抱于脑后，压住双脚以固定双腿。测试时，测试者快速起坐，双肘触及或超过双膝，然后还原为仰卧，双肩胛触及水平面为完成一次，并以此类推累计得出仰卧起坐的连续个数。其中需注意的是，如果测试者借用肘部支撑的力量完成起坐或双肘未触及双膝，该情况不计数。一般地，仰卧起坐的测试时间为1分钟。
基于对动作的规范定义，本发明通过运用图像识别处理的手段实现仰卧起坐的自动识别和计数。为方便说明，本发明以测试者按规范实施动作为前提。
本发明提供如图1所示的一种仰卧起坐的计数方法的流程。
步骤S1中，在测试区域设置具有特定纹路的识别区，识别区为测试时测试者的上半身所在区域，在识别区的斜上方设置摄像机。
请参照如图2，图2是本发明所提供的测试区域示意图。测试区域210包括但并不限于为测试者提供仰卧起坐测试。测试区域210内通过设置识别区211、脚部定位区212和腿部放置区213，如图中黑色虚线框出区域，分别用以标示出测试者测试时上半身、脚部及双腿所应放置的区域。腿部放置区213设置于脚部定位区212和识别区211之间。腿部放置区213不限于与识别区211相接，且腿部放置区213在其宽度W延长方向上较识别区211窄，是为区别双腿与上半身的位置，示意仰卧方向。腿部放置区213内两侧间隔设置有基本沿着测试者仰卧方向延伸的定位线条2131a和定位线条2131b，测试时测 试者的双腿分别沿着所述定位线条2131a和定位线条2131b放置。脚部定位区212设置有固定装置，起到测试时按压双脚以固定双腿的作用。识别区211内部还包括至少一条特定纹路2111，其延伸方向均与测试者仰卧方向基本垂直。本发明特定纹路其图案不受限制，除如图中所示的呈直线条状外，在可选实施例中，特定纹路还可呈均匀分布的曲线、折线、弧线等线状图案，也可以是包括由规则几何图形组成的图案、或是明暗相间的几何图形组合等。
当识别区211内部的特定纹路2111为多条呈如图2所示直线的线状识别图案时，优选的，在测试之前，先获取测试者的身高数据，根据身高数据确定在测试者做出仰卧起坐动作时，针对识别区211内会发生变化的线状识别图案，也就是能够反映测试者上半身运动情况的特定纹路，将其定为关联线状图案。比如不同身高的测试者，仰卧时可能遮挡不同范围的多条特定纹路，当测试者从仰卧到坐立，这些特定纹路会逐渐重现在拍摄的图像中；当测试者由坐立再次恢复到仰卧，这些特定纹路会再逐渐被上半身所遮挡。关联线状图案的动态变化，客观反映测试者的运动过程。此外，关联线状图案的选取，不仅能够在测试过程中，严格要求测试者作动作时须达到仰卧和坐立所需的判断阈值，防止自助测试时的作弊现象，而且，还可以减少数据处理量。
另外，在识别区211的斜上方一侧，通过设置摄像机(图2中未示出)用于对识别区211进行图像拍摄。该摄像机的位置优选设置在与测试者仰卧方向基本垂直的一侧，如图2视野所示意的位置。
步骤S2中，利用摄像机以一定帧率持续拍摄包括测试者和识别区在内的测试图像，得到识别区内的纹路图像。
当测试者在图2所示的测试区域210做好仰卧起坐测试准备时，可用前述 摄像机对测试者持续进行图像拍摄。由于摄像机的拍摄具有特定帧率，因此可得到依时序的若干包括测试者、识别区在内的测试图像。在后续图像处理步骤中将分析识别区内的特定纹路，在定出了前述关联线状图案的情形下可仅为分析关联线状图案即可，即基于关联线状图案在仰卧起坐过程中被拍摄到的长度变化情况，识别出测试者的仰卧起坐完成个数，这将在后面步骤中详述。
步骤S3中，分析得到的纹路图像，当初次发现某帧纹路图像中纹路的像素值达到预设的卧姿阈值时，将该帧设置为起始帧。
能够有效判断仰卧起坐测试的起始帧，则可提高本发明测试方法的准度。因此，在可选实施例中，自前述摄像机开始逐帧拍摄获取测试图像起，实时对测试图像中的关联线状图案进行分析，具体为分析该多条关联线状图案长度的平均值变化。也就是说，测试者在起坐或下躺过程中，上半身会以不同程度遮挡关联线状图案，进而通过摄像机获取的图像中，多条关联线状图案的长度均不同程度地增大或减小，为从整体上分析，将包含在内的多条线状长度的平均值作为分析对象。
在本步骤中，首先预设一卧姿阈值，作为计时起始帧的判断点。卧姿阈值为测试者肩胛骨接触测试区域时摄像机所能拍到的所有关联线状图案长度的平均值。由于在测试图像中，长度具体是由若干像素体现，所以本发明所指线状图案长度，可等同理解为线状图案像素值，而关联线状图案的长度平均值则亦可称为像素平均值。当关联线状图案的长度平均值初次达到预设的卧姿阈值时，视作测试者做好测试准备，并将当前帧设置为起始帧，同时计时开始。
步骤S4中，继续提取分析初始帧后续各帧的纹路图像，每当发现纹路图像中纹路的像素值由预设的卧姿阈值逐渐增大至预设的坐姿阈值时计数一次。
我们知道，一个完整的仰卧起坐是测试者从仰卧姿态到坐立姿态的转变。而本发明中，当测试者从仰卧到坐立，对应地，摄像机持续所获取的关联线状图案的长度平均值则呈现逐渐增大态(多条关联线状图案被测试者上半身遮挡的程度在逐渐减小)。
进一步，测试者起坐后，要求其双肘触及或超过双膝，因此本发明预设一个坐姿阈值，视作测试者坐立的判断点。坐姿阈值为测试者完全坐起时摄像机所能拍到的所有关联线状图案长度的平均值。本步骤中，在自起始帧依序对关联线状图案的处理过程中，关联线状图案在测试图像中的长度平均值会逐渐变大，直至到坐姿阈值。当坐姿阈值满足时，则视作测试者完成了动作转变的要求，则计数一次；若坐姿阈值未满足时，则继续对关联线状图案长度平均值做判断。
如上所述，本发明能够实现对一个仰卧到起坐实现自动启动、识别和计数。而在1分钟测试过程中，对次数的累计则是本方法所述相关步骤的重复再现，即自起始帧以后，基于对关联线状图案平均值的分析，每发现完成了一次坐立动作，则计数次数叠加一次。进一步针对仰卧起坐测试的自动结束，本发明还可以提供方法如下：
一般仰卧起坐要求测试的时间为1分钟。在设置动作起始帧的同时，计时开始，后续持续判定当前帧相对计时开始时刻(起始帧)是否已达到预设时长(如本实施例中的1分钟)，若是，则结束本次测试，即停止提取当前帧测试图像，计时结束，自动保存已累计的仰卧起坐次数；若否，则继续测试。
本发明还提供一种采用前述方法的仰卧起坐计数***，如图3所示的***示意图，其包括：
测试区域210：其包括但并不限于为测试者提供仰卧起坐测试。测试区域210内通过设置具有特定纹路的识别区211、脚部定位区212和腿部放置区213，分别标示出测试者上半身、脚部及双腿所应放置的区域，脚部定位区212还包括设置有固定装置以帮助测试者在测试时固定双腿。三者在测试区域210内的布局与图2所示相同，在此不再赘述。
摄像机220：其位置可参考如图2视野所示意，其设置在识别区211的斜上方，用于对识别区211进行图像拍摄，其位置优选设置在与测试者仰卧方向基本垂直的一侧。摄像机220以一定帧率持续拍摄包括测试者和识别区211在内的测试图像，得到识别区的纹路图像。
数据关联模块230：通过本模块，计数***先在识别区211内直接确定出与测试者相关的特定纹路。具体为，数据关联模块230先根据测试者身高数据确定出测试者做仰卧起坐动作时，通过摄像机220获取的图像中识别区211内会发生长度值变化的多条线状识别图案，将其定为关联线状图案。选取关联线状图案的原因可参照前述关于计数方法的描述中的相关记载，故而在此不再赘述。***后续的模块将分析关联线状图案在仰卧起坐过程中被拍摄到的平均长度值变化情况，客观反映出测试者的运动过程。数据关联模块230的设置并不用于限制本发明计数***。
起始帧获取模块240：其提取并分析摄像机得到的纹路图像，当初次发现某帧纹路图像中纹路的像素值达到预设的卧姿阈值时，将该帧设置为起始帧。具体为在本模块处理中，预设一卧姿阈值，作为计时起始帧的判断点。卧姿阈值为测试者肩胛骨接触测试区域时摄像机所能拍到的所有关联线状图案长度的平均值。当关联线状图案的长度平均值达到预设的卧姿阈值时，视作测试者 做好测试准备，将当前帧设置为起始帧，并生成接续指令，还可以同时发出一起始指令。
纹路图像分析模块250：其继续提取分析起始帧后续各帧的纹路图像，每当发现纹路图像中纹路的像素值由预设的卧姿阈值逐渐增大至预设的坐姿阈值时发出计数指令。具体为在本模块的处理中，预设一坐姿阈值，视作测试者坐立的判断点。坐姿阈值为测试者完全坐起时摄像机所能拍到的所有关联线状图案长度的平均值。纹路图像分析模块250接收由起始帧获取模块240生成的接续指令，继续提取分析摄像机得到的后续各帧的纹路图像，直至发现纹路图像中所有关联线状图案长度平均值由预设的卧姿阈值逐渐增大至预设的坐姿阈值时，就生成一计数指令。
计数模块260：在接收到纹路图像分析模块250生成的计数指令后，对仰卧起坐次数计数一次。
同样，本***能够实现对一个仰卧起坐动作实现自动识别和计数。1分钟测试中，由纹路图像分析模块250持续对关联线状图案长度变化进行分析，作出是否计数的判断。
另外，本***针对测试的自动启动和结束，还可以进一步包括：
计时模块270：其在接收到前述起始帧获取模块240发出的起始指令时开始计时，后续实时记录当前帧相对计时开始时刻的时长数据。
测试结束判定模块280：其调取所述计时模块270记录的时长数据，判定当前帧所对应的时长数据是否已达到预设时长(可为本实施例中的1分钟)，若是，则发出结束指令；否则发出继续指令；所述纹路图像分析模块250在提取分析某帧的纹路图像之前，接收所述测试结束判定模块280发出的指令，以 根据结束指令结束测试，自动保存已累计的仰卧起坐次数，或者根据继续指令继续进行测试。
基于上述方法和***，本发明所实现的仰卧起坐自助测试，可减少人工成本，提高测试效率，给测试者良好的测试体验。
以上所述仅为本发明的较佳实施例而已，并不用以限制本发明，凡在本发明的精神和原则之内所作的任何修改、等同替换和改进等，均应包含在本发明的保护范围之内。

Claims (10)

  1. 一种仰卧起坐测试的计数方法，其特征在于，包括步骤：
    S1、在测试区域设置具有特定纹路的识别区，所述识别区为测试时测试者的上半身所在区域，在所述识别区的斜上方设置摄像机；
    S2、利用所述摄像机以一定帧率持续拍摄包括测试者和识别区在内的测试图像，得到识别区内的纹路图像；
    S3、分析得到的纹路图像，当初次发现某帧纹路图像中纹路的像素值达到预设的卧姿阈值时，将该帧设置为起始帧；
    S4、继续提取分析初始帧后续各帧的纹路图像，每当发现纹路图像中纹路的像素值由预设的卧姿阈值逐渐增大至预设的坐姿阈值时计数一次。
  2. 如权利要求1所述的仰卧起坐的计数方法，其特征在于，步骤S3还包括，在设置动作起始帧的同时，计时开始；步骤S4中，提取某帧的纹路图像之前，先判定该帧所在时刻相对计时开始时刻是否已达到预设时长，是则结束本次测试，否则继续测试。
  3. 如权利要求1所述的仰卧起坐的计数方法，其特征在于，所述识别区内设置的特定纹路为至少一条线状识别图案，所述线状识别图案的延伸方向与测试者仰卧方向基本垂直，各步骤中分析的纹路图像中纹路的像素值为所述线状识别图案的长度值。
  4. 如权利要求3所述的仰卧起坐的计数方法，其特征在于，步骤S1还包括：在测试区域内设定有供测试者脚部放置的脚部定位区，所述脚部定位区与所述识别区之间设置有腿部放置区，所述腿部放置区内间隔设置有基本沿着测试者仰卧方向延伸的两定位线条，测试时测试者的双腿分别沿着所述两定位线条放置；测试前获取测试者的身高数据，根据该身高数据确定出测试者做出仰 卧起坐动作时识别区会发生变化的线状识别图案，定为关联线状图案；步骤S3和S4中分析纹路图像中纹路时，仅分析所述关联线状图案。
  5. 如权利要求4所述的仰卧起坐的计数方法，其特征在于，所述关联线状图案的数量为多条，步骤S3和S4中分析的是所有关联线状图案长度的平均值，所述卧姿阈值为测试者肩胛骨接触测试区域时摄像机所能拍到的所有关联线状图案长度的平均值，所述坐姿阈值为测试者完全坐起时摄像机所能拍到的所有关联线状图案长度的平均值。
  6. 一种仰卧起坐的计数***，其特征在于，包括：
    测试区域，其包括具有特定纹路的识别区，所述识别区为测试时测试者的上半身所在区域；
    摄像机，其设置在所述识别区的斜上方，以一定帧率持续拍摄包括测试者和识别区在内的测试图像，得到识别区的纹路图像；
    起始帧获取模块，其提取并分析摄像机得到的纹路图像，当初次发现某帧纹路图像中纹路的像素值达到预设的卧姿阈值时，将该帧设置为起始帧；
    纹路图像分析模块，其继续提取分析起始帧后续各帧的纹路图像，每当发现纹路图像中纹路的像素值由预设的卧姿阈值逐渐增大至预设的坐姿阈值时发出计数指令；
    计数模块，在接收到纹路图像分析模块发出的计数指令后，对仰卧起坐次数计数一次。
  7. 如权利要求6所述的仰卧起坐的计数***，其特征在于，所述起始帧获取模块在设置起始帧的同时还发出一起始指令，该计数***进一步包括：
    计时模块，其在接收到所述起始帧获取模块发出的起始指令时开始计时， 后续实时记录当前帧相对计时开始时刻的时长数据；
    测试结束判定模块，其调取所述计时模块记录的时长数据，判定当前帧所对应的时长数据否已达到预设时长，是则发出结束指令，否则发出继续指令；所述纹路图像分析模块在提取分析某帧的纹路图像之前，接收所述测试结束判定模块发出的指令，以根据结束指令结束本次测试，或者根据继续指令继续进行测试。
  8. 如权利要求6所述的仰卧起坐的计数***，其特征在于，所述识别区内设置的特定纹路为至少一条线状识别图案，所述线状识别图案的延伸方向与测试者仰卧方向基本垂直，所述起始帧获取模块以及所述纹路图像分析模块分析的纹路图像中纹路的像素值为所述线状识别图案的长度值。
  9. 如权利要求8所述的仰卧起坐的计数***，其特征在于，所述测试区域内还设定有供测试者脚部放置的脚部定位区，所述脚部定位区与所述识别区之间设置有腿部放置区，所述腿部放置区内间隔设置有基本沿着测试者仰卧方向延伸的两定位线条，测试时测试者的双腿分别沿着所述两定位线条放置；
    所述计数***进一步包括数据关联模块，用于测试前获取测试者的身高数据，根据该身高数据确定出测试者做出仰卧起坐动作时识别区会发生变化的线状识别图案，定为关联线状图案；所述起始帧获取模块以及所述纹路图像分析模块分析纹路图像中纹路时，仅分析所述关联线状图案。
  10. 如权利要求9所述的仰卧起坐的计数***，其特征在于，所述关联线状图案的数量为多条，所述起始帧获取模块以及所述纹路图像分析模块分析的是所有关联线状图案长度的平均值，所述卧姿阈值为测试者肩胛骨接触测试区域时摄像机所能拍到的所有关联线状图案长度的平均值，所述坐姿阈值为测试 者完全坐起时摄像机所能拍到的所有关联线状图案长度的平均值。
PCT/CN2016/112276 2016-05-09 2016-12-27 仰卧起坐测试的计数方法及*** WO2017193590A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610298570.2A CN105913045B (zh) 2016-05-09 2016-05-09 仰卧起坐测试的计数方法及***
CN201610298570.2 2016-05-09

Publications (1)

Publication Number Publication Date
WO2017193590A1 true WO2017193590A1 (zh) 2017-11-16

Family

ID=56748546

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/112276 WO2017193590A1 (zh) 2016-05-09 2016-12-27 仰卧起坐测试的计数方法及***

Country Status (2)

Country Link
CN (1) CN105913045B (zh)
WO (1) WO2017193590A1 (zh)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109815907A (zh) * 2019-01-25 2019-05-28 深圳市象形字科技股份有限公司 一种基于计算机视觉技术的仰卧起坐姿态检测与指导方法
CN110732119A (zh) * 2019-10-15 2020-01-31 上海淡竹体育科技有限公司 仰卧起坐测试的方法及装置
CN113011242A (zh) * 2020-12-31 2021-06-22 杭州拓深科技有限公司 一种仰卧起坐计数方法、装置、电子装置和存储介质
CN114209309A (zh) * 2021-12-14 2022-03-22 天津科技大学 一种基于视觉技术的运动行为分析方法
CN118015706A (zh) * 2024-03-21 2024-05-10 广东先知大数据股份有限公司 一种仰卧起坐动作规范检测方法、装置及存储介质

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105913045B (zh) * 2016-05-09 2019-04-16 深圳泰山体育科技股份有限公司 仰卧起坐测试的计数方法及***
CN106650590A (zh) * 2016-09-30 2017-05-10 上海斐讯数据通信技术有限公司 一种仰卧起坐计数方法及装置及智能终端
CN109199324A (zh) * 2017-06-30 2019-01-15 深圳泰山体育科技股份有限公司 基于光学的闭眼单脚站立测量方法及***
CN109200563A (zh) * 2017-06-30 2019-01-15 深圳泰山体育科技股份有限公司 仰卧起坐测试的光学标定及测量方法
CN108887978B (zh) * 2018-07-20 2021-04-06 深圳中云创新技术有限公司 一种坐姿检测***
CN111368810B (zh) * 2020-05-26 2020-08-25 西南交通大学 基于人体及骨骼关键点识别的仰卧起坐检测***及方法
CN113255622B (zh) * 2021-07-14 2021-09-21 北京壹体科技有限公司 一种智能识别仰卧起坐动作姿态完成状况的***和方法

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101172199A (zh) * 2006-07-18 2008-05-07 孙学川 仰卧起坐智能测试***
KR20140015779A (ko) * 2012-07-24 2014-02-07 (주)웰텍 윗몸일으키기 측정장치
CN104667510A (zh) * 2015-02-09 2015-06-03 深圳泰山在线科技有限公司 一种人体动作测试***
CN104688233A (zh) * 2015-02-11 2015-06-10 深圳泰山在线科技有限公司 体质测试机
CN104688237A (zh) * 2015-02-11 2015-06-10 深圳泰山在线科技有限公司 体质检测的测时方法及***
CN105833466A (zh) * 2016-05-25 2016-08-10 深圳市恒康佳业科技有限公司 一种仰卧起坐的测量计数方法及装置
CN105913045A (zh) * 2016-05-09 2016-08-31 深圳泰山体育科技股份有限公司 仰卧起坐测试的计数方法及***

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100255968A1 (en) * 2009-04-07 2010-10-07 Jiang rong-hua Sit-Up Exercising Apparatus
CN203043497U (zh) * 2013-02-01 2013-07-10 张小龙 一种仰卧起坐计数装置
CN204815537U (zh) * 2015-08-17 2015-12-02 温州大学城市学院 一种带有计数器机构的仰卧起坐装置
CN105498166B (zh) * 2016-01-08 2018-08-24 湖南师范大学 一种仰卧起坐自动测试器

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101172199A (zh) * 2006-07-18 2008-05-07 孙学川 仰卧起坐智能测试***
KR20140015779A (ko) * 2012-07-24 2014-02-07 (주)웰텍 윗몸일으키기 측정장치
CN104667510A (zh) * 2015-02-09 2015-06-03 深圳泰山在线科技有限公司 一种人体动作测试***
CN104688233A (zh) * 2015-02-11 2015-06-10 深圳泰山在线科技有限公司 体质测试机
CN104688237A (zh) * 2015-02-11 2015-06-10 深圳泰山在线科技有限公司 体质检测的测时方法及***
CN105913045A (zh) * 2016-05-09 2016-08-31 深圳泰山体育科技股份有限公司 仰卧起坐测试的计数方法及***
CN105833466A (zh) * 2016-05-25 2016-08-10 深圳市恒康佳业科技有限公司 一种仰卧起坐的测量计数方法及装置

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109815907A (zh) * 2019-01-25 2019-05-28 深圳市象形字科技股份有限公司 一种基于计算机视觉技术的仰卧起坐姿态检测与指导方法
CN109815907B (zh) * 2019-01-25 2023-04-07 深圳市象形字科技股份有限公司 一种基于计算机视觉技术的仰卧起坐姿态检测与指导方法
CN110732119A (zh) * 2019-10-15 2020-01-31 上海淡竹体育科技有限公司 仰卧起坐测试的方法及装置
CN113011242A (zh) * 2020-12-31 2021-06-22 杭州拓深科技有限公司 一种仰卧起坐计数方法、装置、电子装置和存储介质
CN114209309A (zh) * 2021-12-14 2022-03-22 天津科技大学 一种基于视觉技术的运动行为分析方法
CN114209309B (zh) * 2021-12-14 2024-06-11 天津市卓越新中新龙腾科技发展有限公司 一种基于视觉技术的运动行为分析方法
CN118015706A (zh) * 2024-03-21 2024-05-10 广东先知大数据股份有限公司 一种仰卧起坐动作规范检测方法、装置及存储介质

Also Published As

Publication number Publication date
CN105913045B (zh) 2019-04-16
CN105913045A (zh) 2016-08-31

Similar Documents

Publication Publication Date Title
WO2017193590A1 (zh) 仰卧起坐测试的计数方法及***
WO2017193591A1 (zh) 俯卧撑测试的计数方法及***
CN111368810B (zh) 基于人体及骨骼关键点识别的仰卧起坐检测***及方法
CN106621284B (zh) 一种智能的体质检测平台
Pal et al. A robust heart rate detection using smart-phone video
WO2017101195A1 (zh) 检测心率的方法和装置
US8391442B2 (en) Radiographic apparatus and imaging method thereof
CN111027417B (zh) 基于人体关键点检测算法的步态识别方法及步态评估***
US20160278683A1 (en) Standing position evaluation apparatus, standing position evaluation method, and non-transitory computer readable medium
WO2013060249A1 (zh) 基于视频图像的非接触式生命体征的检测方法及检测***
CN109999469A (zh) 一种足球绕杆考试智能裁判***
EP2650841A3 (en) Image processing apparatus, image processing method, and image processing program
EP2663230A1 (en) Improved detection of breathing in the bedroom
CN110732119A (zh) 仰卧起坐测试的方法及装置
KR20060066708A (ko) 동잡음 제거된 혈류량 신호 검출 장치 및 방법, 그리고이를 이용한 스트레스 검사 장치
Shukla et al. A comparison of four approaches to evaluate the sit-to-stand movement
KR102369359B1 (ko) 영상기반 지능형 푸시-업 판별 방법 및 푸시-업 판별 시스템
CN116052277A (zh) 一种基于人体检测和骨骼关键点检测的仰卧起坐计数及防作弊***
TW201427646A (zh) 電極貼片模組及酸痛偵測方法及裝置
CN113487635B (zh) 一种基于图像差分的仰卧起坐计数方法
CN106855392A (zh) 煎烤条件分析***与方法
CN106778617B (zh) 一种基于图像采集技术的睡眠监测方法及***
Aldenhoven et al. mBalance: Detect Postural Imbalance with Mobile Devices
CN114022952A (zh) 一种基于视觉技术的高抬腿测试计数方法、装置和介质
CN109200563A (zh) 仰卧起坐测试的光学标定及测量方法

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16901545

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 08.04.2019)

122 Ep: pct application non-entry in european phase

Ref document number: 16901545

Country of ref document: EP

Kind code of ref document: A1