Height adjustment method, apparatus, testing equipment and storage medium

Abstract

This disclosure provides a height adjustment method, apparatus, detection device, and storage medium. The method includes: acquiring a video stream captured by an image acquisition module; performing image processing on image frames of the video stream to obtain height information of a target detection object; acquiring the state of a detection position of the detection device; if the detection position is in an idle state, adjusting the motion module of the detection device to the target height according to the height information; wherein, the idle state means that there is no detection object at the detection position. The method of this disclosure can effectively improve detection efficiency and alleviate the congestion problem of the target object.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and more particularly to a height adjustment method, apparatus, detection device, and storage medium. Background Technology

[0002] Taking morning health check equipment as an example, this equipment can be used for student entry checks. To improve efficiency and accuracy, many kindergartens and primary schools use automated morning health check equipment, which can automatically identify individuals. However, in practice, this type of equipment has poor efficiency, resulting in longer check times and congestion among those waiting to be checked. Summary of the Invention

[0003] This disclosure provides a height adjustment method, apparatus, testing equipment, and storage medium.

[0004] According to a first aspect of this disclosure, a height adjustment method is provided, the method comprising:

[0005] Acquire the video stream captured by the image acquisition module;

[0006] Image processing is performed on the image frames of the video stream to obtain the height information of the target object;

[0007] Obtain the status of the detection position of the detection device;

[0008] If the detection position is in an idle state, the motion module of the detection device is adjusted to the target height according to the height information; wherein, the idle state means that there is no detection object at the detection position.

[0009] In some embodiments, the video stream includes: image frames of target identifiers and / or image frames of people queuing.

[0010] In some embodiments, the image processing of the image frames of the video stream to obtain the height information of the target detection object includes:

[0011] Determine the facial information in the image frame;

[0012] The target facial information of the target face is determined from the facial information; wherein the person corresponding to the target face is the target detection object;

[0013] Based on the target facial information, the height information of the target detection object is determined.

[0014] In some embodiments, determining the target face information from the face information includes:

[0015] Determine the position matrix of multiple faces in the image frame;

[0016] Based on the position matrix, the face with the largest facial area is determined from among the multiple faces as the target face, and the target face information corresponding to the target face is obtained.

[0017] In some embodiments, determining the height information of the target detection object based on the target facial information includes:

[0018] The similarity of the target facial information with facial data in the database is compared to obtain the similarity result;

[0019] If the similarity result is greater than or equal to the face recognition threshold, the historical height information of the target detection object is obtained, and the historical height information is used as the height information.

[0020] In some embodiments, determining the height information of the target detection object based on the target facial information includes:

[0021] If the similarity result is less than the face recognition threshold, the height information of the target object is calculated based on the offset parameter between the position matrix of the target face and the center point of the image frame.

[0022] In some embodiments, the method further includes:

[0023] Determine the image quality of the target face; wherein the image quality includes the image resolution and / or the image area;

[0024] The step of comparing the target facial information with facial data in the database to obtain a similarity result includes:

[0025] If the quality parameter of the image quality is greater than or equal to the parameter threshold, the target facial information is compared with the facial data in the database to obtain the similarity result.

[0026] In some embodiments, the method further includes:

[0027] If the similarity result is less than the face recognition threshold, the recognition information of the target object is recorded in the database; wherein, the recognition information includes the height information.

[0028] In some embodiments, the method further includes:

[0029] Obtain the operating status of the detection equipment;

[0030] If the motion module of the detection device is in a non-motion state and the image processing is not in progress, the video stream acquired by the image acquisition module is acquired again.

[0031] According to a second aspect of this disclosure, a height adjustment device is provided, the device comprising:

[0032] The first acquisition module is used to acquire the video stream acquired by the image acquisition module;

[0033] The image processing module is used to perform image processing on the image frames of the video stream to obtain the height information of the target detection object;

[0034] The second acquisition module is used to acquire the status of the detection position of the detection device.

[0035] An adjustment module is used to adjust the motion module of the detection device to a target height according to the height information when the detection position is in an idle state; wherein, the idle state means that there is no detection object at the detection position.

[0036] According to a third aspect of this disclosure, a detection device is provided, including a memory, a processor, and an executable program stored in the memory and executable by the processor, wherein the processor executes the steps of the method as described in any embodiment of the first aspect when running the executable program.

[0037] According to a fourth aspect of this disclosure, a storage medium is provided having an executable program stored thereon, characterized in that the executable program, when executed by a processor, implements the steps of the method described in any embodiment of the first aspect.

[0038] The height adjustment method and apparatus provided in this disclosure utilize an image acquisition module to acquire a video stream and perform image processing on the image frames of the video stream. This allows for the acquisition of the height information of the target object before it reaches the detection position. If the detection position is idle, the motion module is adjusted to the target height based on the height information. This height adjustment process employs a pre-adjustment mechanism, adjusting the motion module to the target height before the target object is detected. This eliminates the need to wait for the target object to reach the detection position before adjusting the motion module's height, fully utilizing the idle time of the detection equipment during the detection process. This effectively improves detection efficiency and alleviates congestion problems associated with the target object. Attached Figure Description

[0039] Figure 1 This is one of the flowcharts illustrating a height adjustment method according to an exemplary embodiment;

[0040] Figure 2 This is a second schematic flowchart illustrating a height adjustment method according to an exemplary embodiment;

[0041] Figure 3 This is a third flowchart illustrating a height adjustment method according to an exemplary embodiment;

[0042] Figure 4 This is a fourth schematic flowchart illustrating a height adjustment method according to an exemplary embodiment;

[0043] Figure 5 This is a schematic diagram of the structure of a height adjustment device according to an exemplary embodiment. Detailed Implementation

[0044] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses consistent with some aspects of this disclosure as detailed in the appended claims.

[0045] Research has found that morning health screening devices that can automatically adjust camera height are more popular because kindergarten children vary in height. Generally, after the previous child's screening is completed, the device automatically moves its motion module to the center of the track. Once a child stands in the screening position, the motion module moves from top to bottom or bottom to top to locate the child's face. This method involves a relatively long motion module operation time, which is one reason for the low detection efficiency of morning health screening devices.

[0046] The detection equipment in this disclosure includes, but is not limited to, morning inspection equipment.

[0047] To improve the detection efficiency of testing equipment, the present disclosure provides the following technical solutions.

[0048] According to a first aspect of this disclosure, a height adjustment method is provided, such as... Figure 1 As shown, the method includes:

[0049] Step S110: Obtain the video stream acquired by the image acquisition module;

[0050] Step S120: Perform image processing on the image frames of the video stream to obtain the height information of the target detection object;

[0051] Step S130: Obtain the status of the detection position of the detection device;

[0052] Step S140: If the detection position is in an idle state, adjust the motion module of the detection device to the target height according to the height information; wherein, the idle state means that there is no detection object at the detection position.

[0053] In this embodiment, the execution entity of the height adjustment method is the control device of the detection equipment. The control device can be built into the detection equipment as part of it, or it can be external, separate from the detection equipment, and function as an independent electronic device. In this case, the control device can be a computer, server, or mobile phone, etc. An external control device can be connected to the detection equipment wirelessly or via wired communication to achieve communication with the detection equipment.

[0054] In step S110, the image acquisition module can be a camera installed on the detection device, for example, a USB camera (camera using a USB interface) or an IPC (IP CAMERA) camera.

[0055] The image acquisition module can also be a camera device that is set up separately from the detection device.

[0056] A video stream consists of individual image frames. For USB cameras, these frames (frame data) can be obtained through the camera's driver software. For IPC cameras, the frames can be obtained through the SDK (Software Development Kit) provided by the IPC camera manufacturer. After obtaining a frame, the video stream image frame is converted into an image format (usually YUV to JPG) using code. The converted image frame is then used for the next step of image processing.

[0057] In step S120, when the object being measured is a person, the height information refers to body height.

[0058] In some embodiments, image processing includes facial recognition processing. For example, if the detection device is a morning health check device used in a kindergarten, the image processing includes facial recognition processing.

[0059] Without limitation, the height information of the target object can be calculated from the image processing results of the image frame. For example, the processing result includes the coordinates of the target object's face in the image frame. By utilizing the correspondence between the image coordinates in the image frame and the coordinates in physical space, the height information of the target object can be calculated.

[0060] Alternatively, height information can be read directly from the memory of the detection device based on the image processing results of the image frame.

[0061] In steps S130 and S140, the state of the detection bit includes: the detection bit has a detection object, or the detection bit does not have a detection object.

[0062] If a detection target is present at the detection position, it indicates that the detection device is in working condition. Even if the height information of the target object is obtained, the motion module will not be processed. This processing avoids interfering with any ongoing detection.

[0063] If there is no object to be detected at the detection position, it indicates that the detection equipment is in an idle state (also known as a standby state). The idle state may be that the detection equipment has just been turned on and has not yet started detection work; or it may be that the previous object to be detected has just been detected and the next object to be detected (i.e., the target object to be detected) has not yet reached the detection position.

[0064] Unrestricted, such as Figure 3 The detection device 10 shown in Figure a has a track 11 on its outer casing, along which a motion module 20 can slide. A detection camera 40 is mounted on the motion module 20. Driven by motors and other driving components, the motion module 20 can move along the detection track 11 with the detection camera 40. An image acquisition module 30 can be mounted on top of the detection device 10.

[0065] In this embodiment, an image acquisition module acquires a video stream, and image processing is performed on the image frames of the video stream to obtain the height information of the target object before it reaches the detection position. If the detection position is idle, the motion module is adjusted to the target height based on the height information. This height adjustment process employs a pre-adjustment mechanism, adjusting the motion module to the target height before the target object is detected, eliminating the need to wait for the target object to reach the detection position before adjusting the motion module's height. This fully utilizes the idle time of the detection equipment during the detection process, effectively improving detection efficiency and alleviating congestion problems related to the objects to be detected.

[0066] According to some alternative embodiments, the video stream includes: image frames of target identifiers and / or image frames of people queuing.

[0067] Target identification facilitates the identification of target objects. Target identification can be ground markings or wearable items used to identify the target object from other people.

[0068] For example, if the target identifier is a person wearing a yellow hat, then the person wearing the yellow hat is the person to be inspected. Or, if the target identifier is a yellow warning line on the ground, then the queue outside the yellow warning line is the person to be inspected.

[0069] For people in a queue, the target detection object can be the person closest to the image acquisition module, that is, the person at the very front of the queue.

[0070] According to some optional embodiments, the step of performing image processing on the image frames of the video stream to obtain the height information of the target detection object includes:

[0071] Determine the facial information in the image frame;

[0072] The target facial information of the target face is determined from the facial information; wherein the person corresponding to the target face is the target detection object;

[0073] Based on the target facial information, the height information of the target detection object is determined.

[0074] Generally, an image frame includes a face image and a background image. Face recognition models can then be used to identify the face. Typically, an image frame contains multiple faces, meaning it includes multiple facial details. After identifying the face, the target face is then determined from these multiple facial details.

[0075] If only one face is detected in an image frame, that face can be directly identified as the target face.

[0076] Without limitation, facial information includes facial features, the position of the face in the image frame, and other information. Among them, facial features can be used to distinguish and identify faces from background images; position moment information can be used to determine the distance between the corresponding face and the image acquisition module.

[0077] In some embodiments, a large number of candidate images can be trained using a convolutional neural network (lightweight systems such as ARM use the Ultra-Light-Fast-Generic-Face-Detector model, while systems with better performance use the YOLOv5 model) to output a face classification model and a face feature extraction model. The face classification model obtains the position matrix of the face on the image frame, and the face feature extraction model obtains the facial features.

[0078] According to some optional embodiments, determining the target face information from the face information includes:

[0079] Determine the position matrix of multiple faces in the image frame;

[0080] Based on the position matrix, the face with the largest facial area is determined from among the multiple faces as the target face, and the target face information corresponding to the target face is obtained.

[0081] The closer the distance to the image acquisition module, the larger the face area in the image frame. Therefore, the face with the largest face area is the target face.

[0082] In some embodiments, determining the face with the largest facial area from a plurality of faces as the target face based on the position matrix includes:

[0083] Calculate the diagonal distance between the coordinate points at both ends of the diagonal of the position matrix;

[0084] The facial area is determined based on the diagonal distance.

[0085] The greater the diagonal distance, the larger the corresponding face area. The face with the largest diagonal distance is the face closest to the image acquisition module.

[0086] Generally, the coordinates of the two ends of the diagonal of the matrix refer to the XY coordinates of the upper left and lower right points of the corresponding face position matrix, respectively.

[0087] According to some optional embodiments, determining the height information of the target detection object based on the target facial information includes:

[0088] The similarity of the target facial information with facial data in the database is compared to obtain the similarity result;

[0089] If the similarity result is greater than or equal to the face recognition threshold, the historical height information of the target detection object is obtained, and the historical height information is used as the height information.

[0090] Typically, similarity results can be obtained by calculating the variance between the feature matrix of the target face and the facial feature matrix in the database.

[0091] The facial recognition threshold can be 0.80, 0.84, 0.90, etc. The facial recognition threshold can be adjusted according to actual usage.

[0092] If the similarity is greater than or equal to the face recognition threshold, it indicates that the target object is a person already registered in the database. In this case, the height information of the target object can be directly obtained from the database and used as the height information.

[0093] In some embodiments, obtaining the historical height information of the target detection object includes:

[0094] Obtain the most recent historical height information of the target object; or, obtain the historical height information stored in the memory of the detection device.

[0095] Among them, the most recent historical height data will be updated over time, and this height information is more accurate for children who are still growing.

[0096] According to some optional embodiments, determining the height information of the target detection object based on the target facial information includes:

[0097] If the similarity result is less than the face recognition threshold, the height information of the target object is calculated based on the offset parameter between the position matrix of the target face and the center point of the image frame.

[0098] If the similarity result is less than the face recognition threshold, it means that the target object is not a registered person and its height information cannot be obtained from the database. In this case, it is necessary to perform image processing on the image frame to calculate the height information of the target object.

[0099] In some embodiments, adjusting the motion module of the detection device to a target height based on the height information includes:

[0100] Determine the current height of the motion module;

[0101] Based on the current height and the height information, adjust the motion module to the target height.

[0102] Generally, the adjustment direction of the motion module includes moving upwards or downwards. If the difference between the current height and the height indicated by the height information is negative, it indicates that the motion module needs to move upwards a certain distance; conversely, if the difference is positive, it indicates that the motion module needs to move downwards a certain distance. For example: if the current height of the motion module is 100cm, and the height indicated by the height information of the target object is 95cm, then 100-95=5cm, so the motion module needs to be adjusted to move downwards by 5cm. If the height indicated by the height information of the target object is 105cm, then 100-105=-5cm, so the motion module needs to be adjusted to move upwards by 5cm.

[0103] In some embodiments, adjusting the motion module of the detection device to a target height based on the height information includes:

[0104] The movement direction and distance of the motion module are calculated based on the offset parameter between the position matrix of the target face and the center point of the image frame.

[0105] Without limitation, the offset parameters include offset distance and offset direction. The offset distance and offset direction between the position matrix of the target face and the center point of the image frame can be calculated. Based on the proportional relationship between the offset distance and the distance that the motion module should move, the distance that the motion module needs to move is determined; based on the offset direction, the direction that the motion module needs to move is determined.

[0106] According to some optional embodiments, the method further includes:

[0107] Determine the image quality of the target face; wherein the image quality includes the image resolution and / or the image area;

[0108] The step of comparing the target facial information with facial data in the database to obtain a similarity result includes:

[0109] If the quality parameter of the image quality is greater than or equal to the parameter threshold, the target facial information is compared with the facial data in the database to obtain the similarity result.

[0110] Once the image quality parameters of the target face image are determined to be greater than or equal to the parameter threshold, similarity comparison can be performed to effectively distinguish the object to be detected from irrelevant personnel and reduce the useless movement of the motion module.

[0111] If the image quality parameter is less than the parameter threshold, the target facial information will no longer be compared with the facial data in the database, meaning that image processing will not continue.

[0112] For example, if the area of ​​the target face image is less than the area threshold, and / or the resolution of the target face image is less than the resolution threshold, it indicates that the person corresponding to the target face in the image frame is far away from the image acquisition module. This person may only be standing or passing by the detection device and is not the object to be detected. In this case, stopping the processing of such target face images is beneficial to saving computing resources and reducing useless movement of the motion module.

[0113] If the image quality parameter is resolution, then the parameter threshold is the resolution threshold; if the image quality parameter is the area of ​​the image, then the parameter threshold is the area threshold.

[0114] According to some optional embodiments, the method further includes:

[0115] If the similarity result is less than the face recognition threshold, the recognition information of the target object is recorded in the database; wherein, the recognition information includes the height information.

[0116] The identification information may also include the detection result of the target object, the name of the target object, age, etc.

[0117] According to some optional embodiments, the method further includes:

[0118] Obtain the operating status of the detection equipment;

[0119] If the motion module of the detection device is in a non-motion state and the image processing is not in progress, the video stream acquired by the image acquisition module is acquired again.

[0120] In some embodiments, if the motion module of the detection device is in motion, or if image processing is in progress, the video stream acquired by the image acquisition module is not reacquired.

[0121] In this embodiment, the height adjustment process is continuous. For example, the image acquisition module continuously acquires video streams. If the motion module is in motion, or the system is performing image processing, the image frames for the current time period are skipped, and the video stream acquired by the image acquisition module is not reacquired. If the motion module is not in motion, or the system is not performing image processing, image frames are reacquired to move the motion module for the next target detection object. This process reduces computational load and saves resources.

[0122] In a specific example, such as Figure 2 As shown, the height adjustment method includes the following steps:

[0123] Step S210: Acquire the video stream captured by the image acquisition module; wherein, the video stream includes image frames of the people queuing. The image acquisition module can be in a state of continuous video stream acquisition, and the process of the image acquisition module acquiring the video stream is roughly as follows: Figure 3 As shown in a and b. The timing for acquiring the video stream includes when the detection device is in a working state or an idle state. For example, if the detection device is a morning check-up device and the people queuing are children, when a child boards the machine for morning check-up, the video stream captured by the camera (image acquisition module) installed on the top of the detection device is transmitted to the image processing system.

[0124] Step S220: Determine facial information in the image frame. The image frame can be referenced. Figure 3 b in the text.

[0125] Step S230: Determine the position matrix of multiple faces in the image frame.

[0126] Step S240: Based on the position matrix, determine the face with the largest facial area from multiple faces as the target face, and obtain the target face information corresponding to the target face; where the queued person corresponding to the target face is the Nth target detection object; N is a positive integer greater than or equal to 1. The target face can be referenced. Figure 3 c in the text.

[0127] In this example, image processing includes steps S220 to S240. Image processing includes a face detection algorithm and a face segmentation algorithm applied to the image frame. The face detection algorithm can locate faces appearing in the image frame, and the face segmentation algorithm can obtain the face edge contour and the coordinate values ​​of facial key points (used to form a position matrix). The distance between the child and the device is determined by the magnitude of the facial key point coordinate values, and the image of the child's face closest to the device is cropped, while the remaining images are filtered out.

[0128] Step S250: Determine the image resolution of the target face.

[0129] Step S260: If the image resolution is greater than the resolution threshold, compare the similarity between the target face information and the face data in the database to obtain the similarity result.

[0130] Step S270: If the similarity result is greater than the face recognition threshold (the face recognition threshold can be set to 0.84), obtain the historical height information of the Nth target detection object and use the historical height information as the height information.

[0131] Step S280: Obtain the status of the detection position of the detection device.

[0132] Step S290: With the detection position idle, adjust the motion module of the detection device to the target height based on the height information. The target height can be equal to, slightly lower than, or slightly higher than the height indicated by the height information; adjustment can be made as needed. For instructions on adjusting the motion module, please refer to... Figure 3 In the d, the arrow in d indicates the direction of movement of the motion module.

[0133] Once the motion module is moved to the target height, the distance sensor on the detection device will be triggered and a morning check will be performed when the child reaches the detection position.

[0134] Step S300: Determine that the motion module is in a non-motion state and that image processing is not in progress, and then acquire the video stream acquired by the image acquisition module again in order to determine the N+1th target detection object.

[0135] The adjustment method in this example, based on the height records of the motion module in previous morning checks, processes the facial data (image frames) in the video stream captured by the image acquisition module. Before the child's morning check, the motion module is moved automatically to the appropriate height, improving check efficiency and enhancing user experience. Furthermore, the image processing uses a face detection algorithm to extract the rectangular outline of the face, separating it from the background. By filtering the nearest child's face image using facial key point coordinates, it effectively filters out false detections caused by the background and other children's faces.

[0136] In another specific example, such as Figure 4 As shown, the height adjustment method includes the following steps:

[0137] Step S310: Acquire the video stream of the queuing children captured by the image acquisition module; wherein, the image acquisition module can be in a state of continuous video stream acquisition. The timing of acquiring the video stream includes: when the detection device is in a working state or an idle state.

[0138] Step S320: Determine the facial information in the image frame.

[0139] Step S330: Determine the position matrix of multiple faces in the image frame.

[0140] Step S340: Based on the position matrix, determine the face with the largest face area from multiple faces as the target face, and obtain the target face information corresponding to the target face; where the queue person corresponding to the target face is the Nth target detection object; N is a positive integer greater than or equal to 1.

[0141] In this example, image processing includes steps S220 to S240. A large number of candidate images can be trained using a convolutional neural network (lightweight systems such as ARM use the Ultra-Light-Fast-Generic-Face-Detector model, while systems with better performance use the YOLOv5 model) to output a face classification model and a face feature extraction model. The face classification model is used to obtain the position matrix of the face on the image frame (XY coordinates of the top left point and the bottom right point). Based on the position matrix, the position matrix of the largest face in the image frame is taken, and the largest face is taken as the target face.

[0142] Take the position matrix of the largest face in the photo, calculate the direction and distance that the moving module needs to move based on the deviation of the position matrix from the center point in the photo, and send the motor movement command through the serial port;

[0143] Step S350: Compare the target facial information with facial data in the database to obtain the similarity result.

[0144] Step S360: If the similarity result is less than the face recognition threshold (the face recognition threshold can be set to 0.84), calculate the movement direction and movement distance of the motion module based on the offset parameter between the position matrix of the target face and the center point of the image frame.

[0145] Step S370: Obtain the status of the detection position of the detection device.

[0146] Step S380: The detection position is in an idle state. Adjust the motion module to the target height.

[0147] The target face identified in step S340 is cropped using a position matrix. The cropped target face is then transformed to a specified size (determined by the selected model) using OpenCV and input into the feature extraction model to obtain a feature matrix formed by facial features. The obtained facial features are then compared cyclically with the features in the registration database (referring to data blocks) (mainly calculating the variance of the two matrices). The value with the highest similarity is selected. If the highest similarity is greater than the facial recognition threshold, the recognition is considered successful; otherwise, the recognition is considered unsuccessful and the face is not included in the registration database.

[0148] Since the face of the target object is not in the registration library, the direction and distance that the motion module should move are calculated by offset from the center point of the image frame (the ratio of the distance of the face position offset from the center point to the distance the motor should move is adjusted according to the actual resolution). The calculated direction and distance are converted into 16-bit instructions and sent to the motor through the opened serial port. The movement of the motion module is controlled by controlling the motor.

[0149] According to the second aspect of the present disclosure, such as Figure 5 As shown, a height adjustment device 400 is provided, the device comprising:

[0150] The first acquisition module 410 is used to acquire the video stream acquired by the image acquisition module;

[0151] Image processing module 420 is used to perform image processing on the image frames of the video stream to obtain the height information of the target detection object;

[0152] The second acquisition module 430 is used to acquire the status of the detection position of the detection device;

[0153] The adjustment module 440 is used to adjust the motion module of the detection device to a target height according to the height information when the detection position is in an idle state; wherein, the idle state means that there is no detection object at the detection position.

[0154] According to some alternative embodiments, the video stream includes: image frames of target identifiers and / or image frames of people queuing.

[0155] According to some optional embodiments, the image processing module is further configured to:

[0156] Determine the facial information in the image frame;

[0157] The target facial information of the target face is determined from the facial information; wherein the person corresponding to the target face is the target detection object;

[0158] Based on the target facial information, the height information of the target detection object is determined.

[0159] According to some optional embodiments, the image processing module is further configured to:

[0160] Determine the position matrix of multiple faces in the image frame;

[0161] Based on the position matrix, the face with the largest facial area is determined from among the multiple faces as the target face, and the target face information corresponding to the target face is obtained.

[0162] According to some optional embodiments, the image processing module is further configured to:

[0163] The similarity of the target facial information with facial data in the database is compared to obtain the similarity result;

[0164] If the similarity result is greater than or equal to the face recognition threshold, the historical height information of the target detection object is obtained, and the historical height information is used as the height information.

[0165] According to some optional embodiments, the image processing module is further configured to:

[0166] If the similarity result is less than the face recognition threshold, the height information of the target object is calculated based on the offset parameter between the position matrix of the target face and the center point of the image frame.

[0167] According to some alternative embodiments, the apparatus further includes:

[0168] A determining module is used to determine the image quality of the target face; wherein the image quality includes the image resolution and / or the image area;

[0169] The image processing module is also used for:

[0170] If the quality parameter of the image quality is greater than or equal to the parameter threshold, the target facial information is compared with the facial data in the database to obtain the similarity result.

[0171] According to some alternative embodiments, the apparatus further includes:

[0172] An update module is used to record the identification information of the target detection object in the database when the similarity result is less than the face recognition threshold; wherein the identification information includes the height information.

[0173] According to some alternative embodiments, the apparatus further includes:

[0174] The third acquisition module is used to acquire the operating status of the detection device;

[0175] The first acquisition module is further configured to acquire the video stream acquired by the image acquisition module again when the motion module of the detection device is in a non-motion state and the image processing is not in progress.

[0176] According to a third aspect of this disclosure, a detection device is provided, including a memory, a processor, and an executable program stored in the memory and executable by the processor, wherein the processor executes the steps of the method as described in any embodiment of the first aspect when running the executable program.

[0177] According to a fourth aspect of this disclosure, a storage medium is provided having an executable program stored thereon, which, when executed by a processor, implements the steps of the method described in any embodiment of the first aspect.

[0178] The methods disclosed in the several method embodiments provided in this disclosure can be arbitrarily combined without conflict to obtain new method embodiments.

[0179] The features disclosed in the several product embodiments provided in this disclosure can be arbitrarily combined without conflict to obtain new product embodiments.

[0180] The features disclosed in the several method or device embodiments provided in this disclosure can be arbitrarily combined without conflict to obtain new method or device embodiments.

[0181] The above description is merely the preferred embodiment of this disclosure and is not intended to limit the scope of protection of this disclosure. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A height adjustment method, characterized by, The detection equipment has a track on its outer casing, along which a motion module can slide. Driven by a drive unit, the motion module moves along the track. An image acquisition module is mounted on top of the detection equipment. The method includes: Obtain the operating status of the detection equipment; If the motion module of the detection device is in a non-motion state and image processing is not in progress, the video stream acquired by the image acquisition module is obtained; wherein, the video stream includes at least one image frame; Determine the facial information in the image frame; Determine the position matrix of multiple faces in the image frame; Based on the position matrix, the face with the largest facial area is determined from among the multiple faces as the target face, and the target face information corresponding to the target face is obtained; wherein, the person corresponding to the target face is the target detection object; The similarity of the target facial information with facial data in the database is compared to obtain the similarity result; If the similarity result is greater than or equal to the face recognition threshold, the historical height information of the target detection object is obtained, and the historical height information is used as the height information; If the similarity result is less than the face recognition threshold, the height information of the target detection object is calculated based on the offset parameter between the position matrix of the target face and the center point of the image frame; In response to determining the height information, the status of the detection position of the detection device is obtained; If the detection position is in an idle state, the motion module of the detection device is adjusted to the target height according to the height information; wherein, the idle state means that there is no detection object at the detection position.

2. The method according to claim 1, characterized in that, The video stream includes: image frames of the target identifier and / or image frames of the people queuing.

3. The method according to claim 1, characterized in that, The method further includes: Determine the image quality of the target face; wherein the image quality includes the image resolution and / or the image area; The step of comparing the target facial information with facial data in the database to obtain a similarity result includes: If the quality parameter of the image quality is greater than or equal to the parameter threshold, the target facial information is compared with the facial data in the database to obtain the similarity result.

4. The method according to claim 1, characterized in that, The method further includes: If the similarity result is less than the face recognition threshold, the recognition information of the target object is recorded in the database; wherein, the recognition information includes the height information.

5. A height adjustment device, characterized in that, The detection equipment has a track on its outer casing, along which a motion module can slide. Driven by a drive unit, the motion module moves along the track. An image acquisition module is mounted on top of the detection equipment. The device includes: The third acquisition module is used to acquire the operating status of the detection device; The first acquisition module is used to acquire a video stream acquired by the image acquisition module if the motion module of the detection device is in a non-motion state and image processing is not in progress; wherein the video stream includes at least one image frame. An image processing module is used to determine facial information in an image frame; determine a position matrix of multiple faces in the image frame; determine the face with the largest facial area from the multiple faces according to the position matrix, and obtain the target facial information corresponding to the target face; wherein the person corresponding to the target face is the target detection object; compare the target facial information with facial data in a database to obtain a similarity result; if the similarity result is greater than or equal to a facial recognition threshold, obtain the historical height information of the target detection object and use the historical height information as height information; if the similarity result is less than the facial recognition threshold, calculate the height information of the target detection object according to the offset parameter between the position matrix of the target face and the center point of the image frame; The second acquisition module is used to acquire the status of the detection position of the detection device in response to determining the height information; An adjustment module is used to adjust the motion module of the detection device to a target height according to the height information if the detection position is in an idle state; wherein, the idle state means that there is no detection object at the detection position.

6. A detection device, comprising a memory, a processor, and an executable program stored in the memory and executable by the processor, characterized in that, When the processor runs the executable program, it performs the steps of the method as described in any one of claims 1 to 4.

7. A storage medium having an executable program stored thereon, characterized in that, When the executable program is executed by the processor, it implements the steps of the method as described in any one of claims 1 to 4.

Images