Height detection method, system and storage medium for intelligent physical examination machine

Abstract

The embodiment of the present invention discloses a height detection method, system and storage medium for an intelligent physical examination machine, which relates to the field of health detection, and can still accurately measure the height when the video image of the human body is not fully captured, and meets the design requirements of easy use and low cost. The present invention includes: obtaining image information and radar information, wherein the image information is obtained by shooting with a camera, the image information includes an image of the head area of ​​the measured human body, the radar information is obtained after the millimeter wave radar emits millimeter waves, and the radar information includes the lateral distance between the measured human body and the physical examination machine; using the image information to identify a face, and obtaining the coordinates of the highest point of the head of the measured human body according to the identified face; according to the lateral distance between the measured human body and the physical examination machine, the coordinate information of the highest point of the head of the measured human body and the calibration file, the height of the measured human body is obtained.

Description

Technical Field

[0001] The present invention relates to the field of health detection, and in particular to a height detection method, system and storage medium for an intelligent physical examination machine. Background Art

[0002] Height measurement is required in many scenarios, such as hospital physical examinations, security, basic traffic ticket verification, and human body modeling. Although the implementation of height measurement seems simple, it is difficult to simultaneously meet many design requirements such as high accuracy, ease of use, and low cost, and ultimately form a product that is competitive in cost, function, and performance.

[0003] Compared with the traditional measurement method that requires the person to be tested to stand in a fixed position at the physical examination site and measure his height with a ruler, solutions have been proposed to measure human height using ultrasonic detectors, rangefinders or optical imaging. Although these methods are improvements over traditional measurement methods, they all require dedicated sensor equipment, such as ultrasonic detectors, infrared rangefinders, depth sensors, etc., and the final measurement equipment has a single function and can only be used to measure height and weight, resulting in high costs and difficulty in promotion. In addition, due to the outbreak of the epidemic in recent years, it is not advisable to go to public places such as hospitals for measurement. Therefore, it is also necessary to design some solutions that can complete height measurement based on remote unmanned equipment.

[0004] In some current solutions, human body video images are collected remotely, and human height is estimated from face images. This method is relatively simple and fast, and is suitable for crowd monitoring scenarios. It estimates height based on the approximate ratio of face and height. This method is only applicable to situations where people are fixed within the field of view, and the accuracy is not high. There are also some solutions that further extract various feature points of the human body in the image for measurement. The principle based on this is mainly to detect human height based on the principle of invariant proportional relationship in motion projection and the principle of triangle similarity. However, this method requires that the head and feet of the human body are in the field of view of the camera. It is more suitable for height estimation in the security field, and still cannot achieve more accurate human body measurement. If the measurement accuracy is to be further improved, additional sensor equipment, such as millimeter-wave radar, needs to be installed above the human head. However, this installation angle cannot measure vital signs such as heart rate and respiratory rate of the human body. Additional sensor equipment needs to be installed to realize other measurement functions, resulting in low sensor equipment reuse rate and high implementation cost. The intelligent physical examination machine designed using this solution needs to install multiple sets of sensors, which is costly and difficult to effectively reduce the failure rate.

[0005] Some solutions are to make improvements at the software level, such as using the three-dimensional coordinates of skeletal key points to determine the target's height data. However, this solution also has many problems. For example, it is necessary to obtain multiple images of the target object taken from different directions, which results in at least two sets of camera and millimeter-wave radar configurations, which requires additional costs. For example, if the human body is not fully photographed, no matter which two skeletal key points of the human body are obtained, the calculation of height is only an estimate based on the general body proportion relationship, rather than an accurate calculation. Therefore, an image of the entire human body is still required, which is not convenient to use. At least the measured persons are required to stand still one by one for measurement, so the upper limit of the measurement efficiency is not high. There are also problems, such as the need to know the internal parameters of the camera, such as focal length, pixel size, camera position and rotation direction, etc. The amount of data to be processed is large, which requires certain computing resources and hardware conditions, which is not conducive to further reducing costs.

[0006] Therefore, it is necessary to design a method that can still accurately measure height when the video image of the human body is not fully captured, and also needs to take into account the design requirements of ease of use and low cost. Summary of the invention

[0007] The embodiments of the present invention provide a height detection method, system and storage medium for an intelligent physical examination machine, which can accurately measure the height even when the video image of the human body is not fully captured, and meet the design requirements of easy use and low cost.

[0008] To achieve the above object, the embodiments of the present invention adopt the following technical solutions:

[0009] In a first aspect, an embodiment of the present invention provides a method comprising:

[0010] S1. Obtaining image information and radar information, wherein the image information is obtained by taking pictures with a camera, the image information includes an image of the head area of ​​the human body under test, and the radar information is obtained after the millimeter wave radar emits millimeter waves, and the radar information includes the lateral distance between the human body under test and the physical examination machine;

[0011] S2, using the image information to identify a human face, and obtaining the coordinates of the highest point of the head of the measured person according to the identified human face;

[0012] S3. Obtain the height of the person being measured according to the lateral distance between the person being measured and the physical examination machine, the coordinate information of the highest point of the head of the person being measured, and a calibration file.

[0013] In a second aspect, an embodiment of the present invention provides a system, comprising:

[0014] A video acquisition unit, a face detection unit, a head positioning unit, a millimeter wave radar calculation unit, a height calibration unit and a height calculation unit, wherein the video acquisition unit includes a camera, the millimeter wave radar calculation unit includes a millimeter wave radar, the shooting field of view of the camera covers the head area of ​​the measured person, and the emission direction of the millimeter wave signal of the millimeter wave radar is toward the chest area of ​​the measured person; the video acquisition unit is used to obtain image information, wherein the image information is obtained by shooting the one camera, and the image information includes an image of the head area of ​​the measured person;

[0015] The millimeter wave radar calculation unit is used to obtain radar information, wherein the radar information is obtained after the millimeter wave radar transmits the millimeter wave, and the radar information includes the lateral distance between the measured human body and the physical examination machine;

[0016] The face detection unit is used to recognize a face using the image information;

[0017] The head top positioning unit is used to obtain the coordinates of the highest point of the head of the measured person according to the recognized face;

[0018] The height calibration unit is used to write the calibration file into the local memory of the intelligent physical examination machine;

[0019] The height calculation unit is used to obtain the height of the person being measured according to the lateral distance between the person being measured and the physical examination machine, the coordinate information of the highest point of the head of the person being measured and the calibration file.

[0020] In a third aspect, an embodiment of the present invention provides a storage medium storing a computer program or instructions, which, when executed, implements the method in the embodiment.

[0021] The embodiment of the present invention provides a height detection method, system and storage medium for an intelligent physical examination machine. The physical examination machine mentioned in this embodiment uses a camera to collect facial images and a millimeter-wave radar in a dual detection mode, and detects heart rate by combining camera and millimeter-wave radar detection. To detect heart rate, the camera needs to be pointed at the human face area, and to detect heart rate, the millimeter-wave radar needs to be pointed at the human chest. The schematic diagram of the physical examination machine is shown in FIG. Figure 2As shown. The method provided by the present invention mentions that the facial image needs to be clear enough for heart rate detection based on video images, so the distance between the person and the camera should not be too far, so the human body image taken by the physical examination machine at this time cannot capture the whole picture of the human body, and the real height of the human body cannot be accurately calculated by relying on only partial image information of the human body. This embodiment is applied to an intelligent physical examination machine equipped with only one camera and one millimeter-wave radar. It only needs to reuse the camera and the millimeter-wave radar, and no additional modification is required for the intelligent physical examination machine. The purpose of accurately calculating the height of the human body by combining the existing camera on the physical examination machine to collect the upper body image of the human body and the millimeter-wave radar distance measurement is achieved, thereby saving the hardware cost and the modification cost of the intelligent physical examination machine to the greatest extent possible. In general, this embodiment can accurately calculate the height of the human body by combining the millimeter-wave radar to calculate the lateral distance between the person and the physical examination machine and the video image when the human body image collected by the camera is incomplete, so that the height of the human body can also be accurately calculated when the video image in the physical examination machine is incomplete; and there is no need for people to go to the hospital for physical examination, and non-contact physical examination can be performed remotely, which improves the user experience and saves the user's time cost. Furthermore, there is no need to add additional equipment to the smart physical examination machine, saving merchants costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0023] Figure 1 A schematic diagram of the system structure provided by an embodiment of the present invention;

[0024] Figure 2 A schematic diagram of an intelligent physical examination machine provided by an embodiment of the present invention;

[0025] Figure 3 A schematic diagram of calibration using people of known height standing in different places provided by an embodiment of the present invention;

[0026] Figure 4 A schematic diagram of the cross-ratio projection invariance principle of a specific example provided in an embodiment of the present invention;

[0027] Figure 5 In the specific example provided for the embodiment of the present invention, a person whose height is lower than the height of the camera is standing in front of the reference surface is a schematic diagram;

[0028] Figure 6 A schematic diagram of a person whose height is lower than the height of the camera standing behind a reference plane in a specific example provided in an embodiment of the present invention;

[0029] Figure 7 A schematic diagram of a person whose height is higher than the height of the camera standing in front of a reference surface in a specific example provided in an embodiment of the present invention;

[0030] Figure 8 A schematic diagram of a person whose height is higher than the height of the camera standing behind a reference plane in a specific example provided in an embodiment of the present invention;

[0031] Fig. 9 A schematic diagram of a method flow chart provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0032] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The embodiments of the present invention will be described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be interpreted as limiting the present invention. It can be understood by those skilled in the art that, unless specifically stated, the singular forms "one", "one", "said" and "the" used herein may also include plural forms. It should be further understood that the term "including" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or their groups. It should be understood that when we say that an element is "connected" or "coupled" to another element, it can be directly connected or coupled to other elements, or there may also be intermediate elements. In addition, the "connection" or "coupling" used here may include wireless connection or coupling. The term "and / or" used herein includes any unit and all combinations of one or more associated listed items. It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meaning as generally understood by those of ordinary skill in the art to which the present invention belongs. It should also be understood that terms such as those defined in general dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless defined as herein.

[0033] This embodiment can be applied to the common intelligent physical examination machines on the market. Such intelligent physical examination machines are usually equipped with cameras and millimeter-wave radars, and perform height detection based on video images and millimeter-wave radars. The design purpose of this embodiment is to accurately calculate the height of a human body when the video image of the human body is not fully captured and the hardware cost is not increased. That is, a method is designed that requires a method to directly calculate the height even if there is only a half-length photo of the human body on the existing physical examination machine, and no additional cost is required when the detection accuracy is high. This embodiment can be applied to an intelligent physical examination machine that is only equipped with one camera and one millimeter-wave radar, and the camera and millimeter-wave radar on the intelligent physical examination machine are not specifically used for this embodiment, and it can still assume the original function, such as: Compared with traditional detection methods, the detection of vital signs by millimeter-wave radar has the advantages of non-contact, high medical comfort, good patient experience, high resolution, no privacy leakage, low radiation and small size, and has broad application prospects in clinical medicine, family medicine and even epidemic treatment. By specially designing the transmission signal of the millimeter-wave radar and detecting the ups and downs of the human chest, the vital signs of the human body are extracted and processed to obtain physiological characteristics such as breathing signals and heartbeat signals. Therefore, the physical examination method based on video and millimeter-wave radar has very important research significance. This method does not require contact, has fast measurement speed and high efficiency. The intelligent physical examination machine uses a non-contact method based on a combination of video and millimeter-wave radar to detect heart rate and respiratory rate. As an important parameter in physical examination, height detection in the intelligent physical examination machine does not require additional hardware, contact with the instrument, or the person being tested needs to maintain a fixed posture for a long time, and has strong real-time performance. The intelligent physical examination machine uses a non-contact measurement method to detect heart rate and respiratory rate. As an important parameter in physical examination, height detection in the intelligent physical examination machine does not require additional hardware, contact with the instrument, or the person being tested needs to maintain a fixed posture for a long time, and has strong real-time performance. In the smart physical examination machine, in order to measure signals such as heart rate and respiratory rate, the human body cannot be too far away from the camera, and the millimeter-wave radar needs to be installed within the range of ±60 degrees up and down and left and right of the chest. Therefore, the human body image collected by the camera does not include the whole body. The millimeter-wave radar is installed parallel to the chest of the human body. At present, there are many methods for measuring height based on video images and millimeter-wave radar.

[0034] This embodiment is applied to an intelligent physical examination machine equipped with only one camera and one millimeter-wave radar. It only needs to reuse the camera and the millimeter-wave radar, and no additional modification is required for the intelligent physical examination machine. It achieves the purpose of accurately calculating the height of a person by combining the existing camera on the physical examination machine to collect images of the upper body of the person and the millimeter-wave radar distance measurement, thereby saving the hardware cost and the modification cost of the intelligent physical examination machine to the greatest extent possible. Specifically, the embodiment of the present invention provides a height detection method for an intelligent physical examination machine, such as Fig. 9As shown, including:

[0035] S1. Obtain image information and radar information.

[0036] The image information is obtained by taking pictures with a camera, and includes an image of the head area of ​​the person being measured. The radar information is obtained after the millimeter wave radar emits millimeter waves, and includes the lateral distance between the person being measured and the physical examination machine.

[0037] Specifically, when it is detected that the head area of ​​the human body being measured appears in the shooting field of view of the camera, the camera is triggered to capture an image, and the millimeter wave radar is triggered to emit a millimeter wave signal.

[0038] In this embodiment, a camera and a millimeter wave radar are installed on the intelligent physical examination machine, the camera's field of view covers the head area of ​​the human body under test, and the millimeter wave signal of the millimeter wave radar is emitted in the chest area of ​​the human body under test. The calibration file is stored in the local memory of the intelligent physical examination machine.

[0039] S2. Recognize a human face using the image information, and obtain the coordinates of the highest point of the head of the measured person according to the recognized human face.

[0040] S3. Obtain the height of the person being measured according to the lateral distance between the person being measured and the physical examination machine, the coordinate information of the highest point of the head of the person being measured, and a calibration file.

[0041] In the practical application of this embodiment, the millimeter wave radar returns the distance d0 and angle θ between the person and the millimeter wave radar through the time difference between the sensor sending the signal and receiving the signal, and the vertical distance between the person and the physical examination machine is d0cosθ. Millimeter wave is an electromagnetic wave that works between 30GHZ-300GHZ and has a wavelength of millimeter level, which is 1mm-10mm. One advantage of short-wavelength millimeter wave is that it has high accuracy. In the embodiment of the present invention, the millimeter wave radar used for collecting physiological signals and measuring distance is TI's IWR6843 chip. It can emit frequency-modulated continuous waves, which can not only measure the distance of the reflector, but also the speed of the reflector in motion. Since the millimeter wave radar is installed within the range of the chest cavity facing the human body, due to the influence of different human body heights, the millimeter wave radar cannot be completely facing the chest area of ​​each person, but because the millimeter wave radar can detect horizontal ±60 degrees and vertical ±60 degrees, it fully meets the detection requirements. According to the signal parameter estimation principle of the millimeter wave radar, the distance, angle and speed of the radar and the human body are estimated. The millimeter-wave radar uses the time difference between the sensor sending and receiving signals to return the distance between the person and the millimeter-wave radar. Among them, Δt represents the time difference between the sensor sending a signal and receiving a signal, and c is the propagation speed of millimeter waves, that is, the speed of light.

[0042] In this embodiment, in S2, the use of the image information to identify a human face includes: performing grayscale conversion on the image of the head region of the detected human body, and performing face detection using the converted grayscale image.

[0043] Specifically, the grayscale image obtained by conversion is used for face detection, including: using the DNN module of OpenCV to load the Caffe model, and performing face detection to obtain the face position. Perform face tracking, wherein: if face tracking fails, use the goodFeaturesToTrack in OpenCV to extract feature points from the face template to obtain corner points for optical flow estimation. If face tracking is successful, the number of frames frame1 where the face is detected is added by 1. When frame1>N1, calcOpticalFlowPyrLK in OpenCV is used to perform KLT-based feature point tracking to obtain the corner point position after optical flow detection, wherein N1 represents the number of consecutive frames, and head positioning is performed only when the face is detected in N1 consecutive frames. For example: in actual applications, face detection can be performed by first obtaining the current frame image, and then converting the color image into a grayscale image. Then perform face detection, specifically using the DNN module of OpenCV to load the Caffe model to perform face detection to obtain the face position. Then, face tracking is performed, which includes: (1) using goodFeaturesToTrack in opencv to extract feature points from the face template and find the corner points required for optical flow estimation. (2) using calcOpticalFlowPyrLK in opencv to implement KLT-based feature point tracking and obtain the corner point positions after optical flow detection. When tracking fails, go to step (1), otherwise, the number of frames in which the face is detected, frame1, is increased by 1. When frame1>N1, go to step (2) to perform head positioning. Among them, N1 indicates how many consecutive frames the face is detected before the head positioning is performed. In the embodiment of the present invention, N1=10, frame1 indicates the number of frames in which the face is detected. N1=10, frame1 indicates the number of frames in which the face is detected.

[0044] Further, in S2, the step of obtaining the coordinates of the highest point of the head of the measured person according to the recognized face includes:

[0045] The range of the face detected in the image is expanded, and canny edge detection is performed within the expanded range to obtain an edge binary image. Contour extraction is performed on the obtained edge binary image, and then the intersection of the face centerline and the extracted contour is calculated. The vertical coordinate value of the highest point of the head in the image is calculated using the intersection. For example: the face detection image is expanded and then edge detection is performed to determine the highest point of the head, including: 1) within the range of the detected face, the upper, lower, and left sides are expanded to 1 / 4 of the face width. 2) Canny edge detection is performed within the expanded range to obtain an edge binary image. 3) Contour extraction is performed on the edge binary image. 4) The intersection of the face centerline and the contour is calculated. 5) The vertical coordinate of the highest point of the head in the image is calculated based on the intersection.

[0046] Further, in S1, radar information is obtained, including: using the time when the millimeter wave radar sends out the millimeter wave signal and receives the millimeter wave signal, determining that the distance between the measured person and the intelligent physical examination machine is d0. According to the rectangular coordinate system where the measured person is located, the vertical distance between the measured person and the intelligent physical examination machine is obtained. Horizontal distance and longitudinal distance Where θ is the azimuth angle, is the elevation angle. The vertical distance between the human body under test and the intelligent physical examination machine is further obtained as For example: after the first startup of the intelligent medical examination machine, height calibration is required to calibrate the camera and internal parameters on the intelligent medical examination machine, which specifically includes: having a person of known height stand in different places and repeat steps (1), (2), and (3) three times to obtain the distances d1, d2, and d3 between the person and the medical examination machine, and the vertical coordinates y1, y2, and y3 of the highest point of the head in the image. After completing the height calibration, it can be used directly without repeating the height calibration in each measurement. It only needs to be calibrated again in the subsequent regular maintenance to facilitate parameter calibration.

[0047] In this embodiment, it also includes: performing height calibration after the intelligent physical examination machine is first started, and writing the result of the height calibration into the calibration file, wherein the calibration file includes: the lateral distance between the intelligent physical examination machine and the reference surface, the projection height of a test person with known height at different positions on the reference surface, and the vertical coordinate of the highest point of the test person at the different positions. Specifically, in this embodiment, the triangle similarity principle can be used to calculate the projection height of a person with known height at different positions on the reference surface. For example: Figure 3 As shown, assume that the camera installation point is C, the camera vertical ground point is E, and the height of the camera from the ground is h c, the height of the known person is h during initial calibration. The plane where the person with known height is located at the first position point F is set as the reference plane. The projection of the head position of the person standing at the first position point F in the reference plane is H1 (calibration point 1), then H1F = h, the distance from the first position point to the physical examination machine is EF = d1 calculated by millimeter wave radar, the projection of the head position of the person standing at the second position point T on the reference plane is H2 (calibration point 2), the distance from the second position point to the physical examination machine is ET = d2 calculated by millimeter wave radar, the projection of the head position of the person standing at the third position point S on the reference plane is H3 (calibration point 3), and the distance from the second position point to the physical examination machine is ES = d3 calculated by millimeter wave radar.

[0048] According to the triangle similarity principle ( Figure 3 ), Calculation shows that:

[0049]

[0050] According to the triangle similarity principle ( Figure 3 ), Calculation shows that:

[0051]

[0052] Write the horizontal distance d1 between the physical examination machine and the reference surface, the projection heights h1, h2, h3 of the head positions of the three positions on the reference surface, and the vertical coordinates y1, y2, y3 of the highest point of the person in the video image into the height calibration file. Assume that the three calibration points in the reference surface are Z1, Z2, and Z3 in the imaging planes H1, H2, and H3, respectively, and the coordinates of the three points in the imaging plane are (x1, y1), (x2, y2), and (x3, y3), respectively. Unify the horizontal coordinates of the three points in the imaging plane, that is, x1 = x2 = x3, so that the three points are collinear in a straight line, and the projection heights H1F, H2F, and H3F of the head positions of the three positions on the reference surface are recorded as h1, h2, and h3.

[0053] In actual application, the process of calculating the height of the person being measured includes the following: 1. Face detection: When someone stands in front of the physical examination machine, height detection starts, and face detection is performed using the same steps as in step one. 2. Head positioning: Use the same steps as in step one to position the head. Determine the highest point H' of the human head in the video image, and obtain the head image coordinates (x4, y4). Let x4=x1=x2=x3, then H' is collinear with Z1, Z2, and Z3. 3. Based on the cross-ratio projection invariance principle, the height of the highest point of the human head in the reference plane from the ground can be calculated. For example:

[0054] The cross-ratio projection invariance principle is as follows Figure 4 As shown in the figure, if there are four points A, B, C, and D collinearly on the line L1, they are projected onto the line L2 through the projection center, and the corresponding image points are A', B', C', and D', then the following relationship exists:

[0055]

[0056] The relationship between the highest point H of the human body in the reference plane and the calibration points H1, H2, H3 on the reference plane and the imaging plane satisfies the cross-ratio projection invariance principle, and the height of the head on the reference plane can be calculated based on this relationship.

[0057] When H' is above Z2, that is, y4 < y2, the points H, H1, H2, H3 on the reference plane and the points H', Z1, Z2, Z3 on the imaging plane satisfy the following proportional relationship:

[0058]

[0059] Among them, distance() represents the distance between two points.

[0060] Then formula (4) becomes:

[0061]

[0062] Let abs(y2-y4)×abs(y3-y1) be represented by parameter α, and abs(y2-y1)×abs(y3-y4) be represented by parameter β, then:

[0063]

[0064] When H' is exactly at point Z2, that is, y4=y2, then h4=h2.

[0065] When H' is below Z2, that is, y4>y2, the points H, H1, H2, H3 on the reference plane and the points H', Z1, Z2, Z3 on the imaging plane satisfy the following proportional relationship:

[0066]

[0067] Among them, distance() represents the distance between two points.

[0068] Then formula (7) becomes:

[0069]

[0070] Express abs(y4-y1)×abs(y3-y2) by parameter δ, then:

[0071]

[0072] Among them, the lateral distance d4 between the human body and the medical examination machine is returned by the millimeter-wave radar. The actual human height is calculated according to the principle of similar triangles. For example Figure 5 , as shown in Figures 6, 7, and 8, the MP line segment represented by the blue line represents the human body to be detected.

[0073] (1) When the height is lower than the camera, that is, h4 < h c .

[0074] 1) When the person stands in front of the reference plane, that is, d4 < d1, as Figure 5 shown, according to the principle of similar triangles, there is That is:

[0075] Among them, CG = h c - GE = h c - HF = h c - h4, GH represents the distance between the medical examination machine and the reference plane, that is, GH = d1, MP = height, KP = HF = h4, GK = EP = d4.

[0076] Therefore, there is

[0077] The actual human height height can be calculated by the above formula, and its calculation formula is as follows:

[0078]

[0079] 2) When the person stands behind the reference plane, that is, d4 > d1, as Figure 6 shown, according to the principle of similar triangles, there is That is:

[0080] Among them, CG = h c - GE = h c - height, GK represents the distance between the medical examination machine and the reference plane, that is, GK = d1, HF = h4, KF = MP = height, GM = d4

[0081] Therefore, there is

[0082] The actual human height height can be calculated by the above formula, and its calculation formula is as follows:

[0083]

[0084] 3) When the person stands on the reference plane, that is, d4 = d1, height = h4.

[0085] (2) When the height is equal to the height of the camera, h4 = h c , then height = h4.

[0086] (3) When the height is higher than the height of the camera, i.e., h4 > h c .

[0087] 1) When the person stands in front of the reference plane, i.e., d4 < d1, as Figure 7 shown, according to the principle of similar triangles, there is i.e.: Then:

[0088]

[0089] 2) When the person stands behind the reference plane, i.e., d4 > d1, as Figure 8 shown, according to the principle of similar triangles, there is i.e.: Then:

[0090]

[0091] 3) When the person stands at the reference plane, i.e., d4 = d1, height = h4.

[0092] Specifically, when the height is higher than the height of the camera, the closer the person is to the camera, the higher the position of the person's head in the video image, and the farther the person is from the camera, the lower the position of the person's head in the video image. When the height is lower than the height of the camera, the closer the person is to the camera, the lower the position of the person's head in the video image, and the farther the person is from the camera, the higher the position of the person's head in the video image. Based on this, it can be moved back and forth to make the human face within the range of the video image.

[0093] In this embodiment, the medical examination machine mentioned uses a dual-detection mode of collecting facial images through a camera and combining with a millimeter-wave radar to detect the heart rate. For the camera to detect the heart rate, the camera needs to face the human facial area, and for the millimeter-wave radar to detect the heart rate, the millimeter-wave radar needs to face the position of the human chest. The schematic diagram of the medical examination machine is as Figure 2As shown. The heart rate detection based on video images mentioned in the method provided by the present invention requires that the facial image is clear enough, so the distance between the person and the camera should not be too far. Therefore, the human body image taken by the physical examination machine at this time cannot capture the whole picture of the human body, and the real height of the human body cannot be accurately calculated based on only partial image information of the human body. In general, the advantages of this embodiment are: when the human body image collected by the camera is incomplete, the millimeter wave radar can be used to calculate the lateral distance between the person and the physical examination machine and the video image to accurately calculate the height of the human body, so that when the video image in the physical examination machine is incomplete, the height of the human body can also be accurately calculated; and there is no need for people to go to the hospital for a physical examination, and non-contact physical examinations can be performed remotely, which improves the user experience and saves the user's time cost. Furthermore, there is no need to add additional equipment to the smart physical examination machine, saving the merchant's cost.

[0094] This embodiment also provides a height detection system for an intelligent physical examination machine, such as Figure 1 As shown, the system includes: a video acquisition unit, a face detection unit, a head positioning unit, a millimeter wave radar calculation unit, a height calibration unit and a height calculation unit, wherein the video acquisition unit includes a camera, the millimeter wave radar calculation unit includes a millimeter wave radar, the camera's shooting field of view covers the head area of ​​the measured person, and the millimeter wave signal of the millimeter wave radar is emitted in the direction of the chest area of ​​the measured person. The height calibration unit is used to write the calibration file into the local memory of the intelligent physical examination machine.

[0095] The video acquisition unit is used to obtain image information, wherein the image information is captured by the one camera, and the image information includes an image of the head area of ​​the measured human body.

[0096] The millimeter wave radar calculation unit is used to obtain radar information, wherein the radar information is obtained after the one millimeter wave radar transmits the millimeter wave, and the radar information includes the lateral distance between the measured human body and the physical examination machine.

[0097] The face detection unit is used to recognize faces using the image information.

[0098] The head top positioning unit is used to obtain the coordinates of the highest point of the head of the measured person according to the recognized face.

[0099] The height calculation unit is used to obtain the height of the person being measured according to the lateral distance between the person being measured and the physical examination machine, the coordinate information of the highest point of the head of the person being measured and the calibration file.

[0100] Specifically, the main function of the video acquisition unit is to shoot the required scene through a general analog camera and obtain analog video images, and then convert them into digital image data through a general video acquisition card. There are certain requirements for the installation height and angle of the camera, and the shooting field of view. The camera should be installed so that when the human body stands in front of the physical examination machine, the head area of ​​the human body appears in the video screen. The camera should be set up so that the front face information of the human face can be clearly displayed in the video screen. It is best to face the human face directly, so the camera is required to shoot from a close distance.

[0101] Face detection unit: The main function of the face detection unit is to convert the input color digital image data into a grayscale image, and then perform face detection on the grayscale image. In order to improve the detection efficiency, after the face is detected in the first frame, the face detection range of the subsequent frames is detected within the expanded range of the face in the previous frame.

[0102] Head positioning unit: The main function of the head positioning unit is to determine the range of head positioning according to the face detected by the face detection unit, and perform contour detection within the range to locate the highest point of the human head.

[0103] Millimeter-wave radar calculation unit: When the face detection unit detects a face image, the millimeter-wave radar is used to detect the lateral distance between the person and the physical examination machine.

[0104] Height calibration unit: A person of known height stands in three different places. The millimeter-wave radar automatically obtains the lateral distance between the person and the physical examination machine. Based on the principle of triangle similarity, the projection height of the person of known height at different positions on the reference surface is obtained. The lateral distance between the physical examination machine and the reference surface, the projection height of the head position at three positions on the reference surface, and the vertical coordinate of the highest point of the person in the video image are written into the height calibration file.

[0105] Height calculation unit: The height is calculated based on the position of the highest point of the head in the video image located by the head positioning unit, the lateral distance between the person and the physical examination machine calculated by the millimeter wave radar calculation unit, and the information in the calibration file.

[0106] An embodiment of the present invention further provides a storage medium storing a computer program or instruction. When the computer program or instruction is executed, the method in the embodiment is implemented.

[0107] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The above is only a specific implementation method of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or replacements that can be easily thought of by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A height detection method for an intelligent physical examination machine, characterized in that: include: S1. Obtaining image information and radar information, wherein the image information is obtained by taking pictures with a camera, the image information includes an image of the head area of ​​the human body under test, and the radar information is obtained after the millimeter wave radar emits millimeter waves, and the radar information includes the lateral distance between the human body under test and the intelligent physical examination machine; S2, using the image information to identify a human face, and obtaining the coordinates of the highest point of the head of the measured person according to the identified human face; S3, obtaining the height of the person being measured according to the lateral distance between the person being measured and the intelligent physical examination machine, the coordinate information of the highest point of the head of the person being measured and the calibration file; It also includes: performing height calibration after the intelligent physical examination machine is started for the first time, and writing the result of the height calibration into the calibration file, wherein the calibration file includes: the lateral distance between the intelligent physical examination machine and a reference surface, the projection height of a test person with known height at different positions on the reference surface, and the vertical coordinate of the highest point of the test person at the different positions.

2. The method according to claim 1, characterized in that Also includes: When it is detected that the head area of ​​the human body being measured appears in the shooting field of view of the camera, the camera is triggered to capture an image, and the millimeter wave radar is triggered to emit a millimeter wave signal.

3. The method according to claim 1, characterized in that A camera and a millimeter-wave radar are installed on the intelligent physical examination machine, the shooting field of the camera covers the head area of ​​the human body under test, and the emission direction of the millimeter-wave signal of the millimeter-wave radar is toward the chest area of ​​the human body under test; The calibration file is stored in the local memory of the intelligent physical examination machine.

4. The method according to claim 1, characterized in that: In S2, the step of recognizing a face using the image information includes: Performing grayscale conversion on the image of the head area of ​​the measured human body; The converted grayscale image is used for face detection.

5. The method according to claim 4, characterized in that The method of performing face detection using the grayscale image obtained by conversion includes: Use the DNN module of OpenCV to load the Caffe model and perform face detection to get the face position; Perform face tracking, where: if face tracking fails, use goodFeaturesToTrack in opencv to extract feature points from the face template to obtain corner points for optical flow estimation; If face tracking is successful, the number of frames where the face is detected is counted. Add 1, when When , calcOpticalFlowPyrLK in opencv is used to track feature points based on KLT to obtain the corner point positions after optical flow detection, where: Indicates the number of consecutive frames. Head location is performed only when a face is detected in the frame.

6. The method according to claim 1 or 4, characterized in that: In S2, the step of obtaining the coordinates of the highest point of the head of the measured person according to the recognized face includes: Expand the range where the face is detected in the image, and perform canny edge detection in the expanded range to obtain an edge binary image; Perform contour extraction on the obtained edge binary image, and then calculate the intersection point between the face center line and the extracted contour; The intersection point is used to calculate the vertical coordinate value of the highest point of the head in the image.

7. The method according to claim 1, characterized in that In S1, radar information is obtained, including: The distance between the human body under test and the intelligent physical examination machine is determined by using the time when the millimeter wave radar sends out the millimeter wave signal and receives the millimeter wave signal. ; According to the rectangular coordinate system where the human body under test is located, the vertical distance between the human body under test and the intelligent physical examination machine is obtained. , Horizontal distance and longitudinal distance ,in, is the azimuth, is the elevation angle.

8. A height detection system for an intelligent physical examination machine, characterized in that: The system comprises: a video acquisition unit, a face detection unit, a head positioning unit, a millimeter wave radar calculation unit, a height calibration unit and a height calculation unit, wherein the video acquisition unit comprises a camera, the millimeter wave radar calculation unit comprises a millimeter wave radar, the shooting field of the camera covers the head area of ​​the measured person, and the emission direction of the millimeter wave signal of the millimeter wave radar is toward the chest area of ​​the measured person; the video acquisition unit is used to obtain image information, wherein the image information is obtained by shooting the one camera, and the image information includes an image of the head area of ​​the measured person; The millimeter wave radar calculation unit is used to obtain radar information, wherein the radar information is obtained after the millimeter wave radar transmits the millimeter wave, and the radar information includes the lateral distance between the human body under test and the intelligent physical examination machine; The face detection unit is used to recognize a face using the image information; The head top positioning unit is used to obtain the coordinates of the highest point of the head of the measured person according to the recognized face; The height calibration unit is used to write the calibration file into the local memory of the intelligent physical examination machine; The height calculation unit is used to obtain the height of the person being measured according to the lateral distance between the person being measured and the intelligent physical examination machine, the coordinate information of the highest point of the head of the person being measured and the calibration file; It also includes: performing height calibration after the intelligent physical examination machine is started for the first time, and writing the result of the height calibration into the calibration file, wherein the calibration file includes: the lateral distance between the intelligent physical examination machine and a reference surface, the projection height of a test person with known height at different positions on the reference surface, and the vertical coordinate of the highest point of the test person at the different positions.

9. A storage medium, characterized in that: A computer program or instruction is stored, and when the computer program or instruction is executed, the method according to any one of claims 1 to 7 is implemented.

Images