Personnel health detection method and device and electronic equipment

Abstract

The invention is suitable for the technical field of computers, and provides a personnel health detection method and device and electronic equipment.The method comprises the steps that a video clip of a to-be-detected person is obtained, and the video clip comprises at least two video frames; video frames of the video clip are preprocessed, a preprocessed video frame sequence is obtained, and preprocessing comprises elimination of high-frequency noise and low-frequency illumination fluctuation of the video frames; and determining the physiological parameters of the to-be-tested person according to the preprocessed video frame sequence. By means of the method, the accuracy of the determined physiological parameters of the to-be-tested person can be improved.

Description

Technical Field

[0001] This application belongs to the field of computer technology, and in particular relates to methods, devices, electronic equipment, computer-readable storage media, and computer program products for monitoring the health of personnel. Background Technology

[0002] As living standards improve, people are paying increasing attention to their health. In daily life, people may regularly go to the hospital for physical examinations to detect any potential health problems. Similarly, in the workplace, considering that individual health risks could increase the risk of accidents, companies may also conduct health checks on their employees. For example, considering that operators handle equipment, and the quality of their operations directly impacts their safety and even the safety of the entire company, health checks can be prioritized for operators. Taking a nuclear power plant as an example, analysis of numerous human-caused incidents on-site shows that the main cause is often operator-related. Therefore, to reduce the probability of human-caused incidents at nuclear power plants, health checks can be conducted on their operators.

[0003] Since physiological parameters (such as heart rate) can reflect a person's health to a certain extent, these physiological parameters can be tested when conducting health checks on individuals.

[0004] Existing methods utilize contact-based detection devices to measure physiological parameters, such as pulse oximeters and blood pressure monitors to measure heart rate. However, this method suffers from low efficiency due to the contact between the devices and the human body. In large organizations, such as nuclear power plants, especially during major overhauls, the sheer number of participants (5,000-20,000 people can enter and exit the overhaul site daily) means that low-efficiency personnel monitoring would severely impact plant operations. Conversely, neglecting personnel monitoring could increase the probability of accidents due to the varying skill levels of the staff.

[0005] To reduce the probability of safety accidents caused by human factors, a detection method with high detection efficiency can be used to detect the suitability of the personnel participating in the maintenance on the same day. Specifically, supervision and intervention should be carried out when the personnel enter the site and during the work process to prevent human factors from failing.

[0006] Compared to detection devices that require contact with the human body to measure physiological parameters, non-contact methods offer higher efficiency. For example, remote photoplethysmography (rPPG) can be used to calculate physiological parameters. rPPG is a non-contact physiological signal monitoring technology that uses a camera to capture subtle color changes on the skin surface to measure physiological parameters such as heart rate, respiratory rate, blood oxygen saturation, and heart rate variability. However, this method may result in lower accuracy of the measured results. Summary of the Invention

[0007] This application provides a method, apparatus, and electronic device for detecting human health, which can solve the problem that the accuracy of detection results obtained when detecting human physiological parameters through non-contact methods is low.

[0008] In a first aspect, embodiments of this application provide a method for detecting the health of a person, including: Acquire a video clip of the person to be tested, wherein the video clip includes at least two video frames; The video frames of the video segment are preprocessed to obtain a preprocessed video frame sequence, wherein the preprocessing includes eliminating high-frequency noise and low-frequency illumination fluctuations in the video frames; The physiological parameters of the person being tested are determined based on the preprocessed video frame sequence.

[0009] The beneficial effects of the embodiments of this application compared with the prior art are: In this embodiment, after acquiring the video segment of the person to be tested, the video frames of the video segment are preprocessed. Since the preprocessing includes eliminating high-frequency noise and low-frequency illumination fluctuations in the video frames, the high-frequency noise and low-frequency illumination fluctuations in the preprocessed video frame sequence will be less than those in the unprocessed video frame sequence. Therefore, when determining the physiological parameters of the person to be tested based on the preprocessed video frame sequence, the interference of high-frequency noise and low-frequency illumination in determining the physiological parameters can be reduced, thereby improving the accuracy of the determined physiological parameters of the person to be tested.

[0010] Optionally, determining the physiological parameters of the person under test based on the preprocessed video frame sequence includes: The physiological parameters of the person under test are determined based on the preprocessed video frame sequence and vibration imaging technology.

[0011] Optionally, determining the physiological parameters of the person under test based on the preprocessed video frame sequence and vibration imaging technology includes: The micro-motion characteristics of the preprocessed video frame sequence are determined based on vibration imaging technology; Enhance the micro-motion features of the skin in the preprocessed video frame sequence to obtain an enhanced video frame sequence; The physiological parameters of the person being tested are determined based on the enhanced video frame sequence.

[0012] Optionally, enhancing the micro-motion features of the skin in the preprocessed video frame sequence to obtain the enhanced video frame sequence includes: When the preprocessed video frame sequence indicates that the person under test is in motion, the micro-motion features of the skin in the preprocessed video frame sequence are enhanced to obtain an enhanced video frame sequence.

[0013] Optionally, the health detection method for the personnel further includes: When the preprocessed video frame sequence indicates that the person under test is in a static state, the physiological parameters of the person under test are determined based on the preprocessed video frame sequence.

[0014] Optionally, enhancing the micro-motion features of the skin in the preprocessed video frame sequence to obtain the enhanced video frame sequence includes: The following steps are performed on the video frames in the preprocessed video frame sequence: the video frame is decomposed into a pyramid to obtain the decomposed image frame sequence corresponding to the video frame; the detailed information of the video frame is determined based on the decomposed image frame sequence corresponding to the video frame; the frequency component corresponding to the heartbeat in the detailed information of the video frame is extracted to obtain the heartbeat component; the heartbeat component is amplified; and the video frame is reconstructed based on the amplified heartbeat component and the decomposed image frame sequence corresponding to the video frame to obtain the reconstructed video frame. An enhanced video frame sequence is obtained based on each of the reconstructed video frames.

[0015] Optionally, determining the physiological parameters of the person under test based on the enhanced video frame sequence includes: Calculate the mean of the enhanced video frame sequence across different channels; Construct a mean function for each channel based on the mean value of each channel; The physiological parameters of the test subject are determined based on each of the aforementioned mean functions.

[0016] Optionally, determining the physiological parameters of the subject based on each of the mean functions includes: The heart rate of the person being tested is determined based on each of the aforementioned mean functions; The heart rate variability index of the subject is calculated based on the heart rate.

[0017] Optionally, calculating the heart rate variability index of the subject based on the heart rate includes: When the duration of the video segment exceeds a preset duration threshold, the heart rate variability index of the person being tested is calculated based on the heart rate. When the duration of the video segment is not greater than the duration threshold, the heart rate variability index to be corrected for the person under test is calculated based on the heart rate, the heartbeat of the person under test is calculated based on the heart rate, and the heart rate variability index to be corrected is corrected based on the heartbeat to obtain the final heart rate variability index of the person under test.

[0018] Optionally, after determining the physiological parameters of the person under test based on the preprocessed video frame sequence, the method further includes: Determine whether the physiological parameters of the person being tested match the warning conditions corresponding to the preset warning level; If the physiological parameters of the person being tested match the warning conditions corresponding to the preset warning level, then the warning corresponding to the matching warning conditions is triggered.

[0019] Optionally, the step of triggering an alert corresponding to the matching alert condition if the physiological parameters of the person being tested match the alert conditions corresponding to a preset alert level includes: If the physiological parameters of the person being tested match the physiological parameters corresponding to the preset warning level, and the duration matches the duration corresponding to the preset warning level, then a warning corresponding to the matching warning conditions is triggered, wherein the warning conditions corresponding to the preset warning level include physiological parameters and duration.

[0020] Optionally, before determining whether the physiological parameters of the person to be tested match the warning conditions corresponding to the preset warning level, the method further includes: Obtain the current ambient light intensity and / or the job type of the person being tested; The warning conditions are determined based on the ambient light intensity and / or the job type.

[0021] Optionally, the preset warning levels include Level 1, Level 2, and Level 3 warnings, with Level 3 being the highest. Following the warning corresponding to the triggered and matched warning conditions, the system further includes: If the warning level is higher than the first-level warning, the video clip of the person to be tested is reacquired, and the physiological parameters of the person to be tested are determined based on the reacquired video clip.

[0022] Optionally, the step of reacquiring the video clip of the person to be tested includes: The video segment of the person under test is reacquired using a target frame rate, wherein the target frame rate is higher than the frame rate corresponding to the first acquisition of the video segment of the person under test.

[0023] Secondly, embodiments of this application provide a health detection device for personnel, comprising: The video clip acquisition module is used to acquire video clips of the person to be tested, wherein the video clips include at least two video frames; A preprocessing module is used to preprocess the video frames of the video segment to obtain a preprocessed video frame sequence, wherein the preprocessing includes eliminating high-frequency noise and low-frequency illumination fluctuations in the video frames; The physiological parameter determination module is used to determine the physiological parameters of the person to be tested based on the preprocessed video frame sequence.

[0024] Thirdly, embodiments of this application provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method as described in any of the first aspects.

[0025] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in any of the first aspects.

[0026] Fifthly, embodiments of this application provide a computer program product that, when run on an electronic device, causes the electronic device to execute the method described in the first aspect above.

[0027] It is understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0029] Figure 1 This is a schematic flowchart of a method for health detection of personnel provided in an embodiment of this application; Figure 2 This is a schematic diagram of the structure of a health monitoring device for personnel provided in one embodiment of this application; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0030] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0031] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0032] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0033] References to "one embodiment" or "some embodiments" in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized.

[0034] Currently, in order to reduce the probability of safety accidents caused by human factors, health checks are usually carried out on personnel. For example, when personnel are nuclear power plant workers (such as maintenance personnel), health checks are carried out on them when they enter the site and during the work process to prevent human-caused failures.

[0035] To improve detection speed, non-contact methods are often chosen for health checks. For example, rPPG (rhapsandometric pulsed light) is used to detect heart rate. rPPG is a non-contact physiological signal monitoring technology that infers changes in blood volume by capturing minute color changes on the skin surface, and then calculates the user's heart rate. However, because fluctuations in ambient light can introduce significant artifacts, the accuracy of heart rate determination using rPPG may be relatively low.

[0036] To improve the accuracy of physiological parameters of individuals being tested, this application provides a method for health detection of individuals.

[0037] The method for health monitoring of personnel provided in the embodiments of this application is described below with reference to the accompanying drawings.

[0038] Figure 1 A flowchart illustrating a method for health detection of personnel according to an embodiment of this application is shown. This method can be applied to electronic devices, and is described in detail below: S11, acquire a video clip of the person to be tested, the video clip including at least two video frames.

[0039] The aforementioned video clips can be video clips acquired during the movement of the person being tested, or video clips acquired when the person being tested is stationary.

[0040] The number of people to be tested can be equal to 1 or greater than 1. For example, when a video clip involves only one human face, the number of people to be tested in that video clip is equal to 1; while when a video clip includes two or more human faces, the number of people to be tested in that video clip is greater than 1.

[0041] S12, preprocess the video frames of the above video segment to obtain a preprocessed video frame sequence, wherein the preprocessing includes reducing high-frequency noise and low-frequency illumination fluctuations in the above video frames.

[0042] Specifically, face detection can be performed on video frames within the video clip, using tools such as MTCNN or MediaPipe Face Mesh. MTCNN (Multi-Task Cascaded Convolutional Networks) is a deep learning method for face detection and keypoint localization, while MediaPipe Face Mesh is a machine learning-based facial landmark detection tool developed by Google, capable of accurately locating 468 3D facial landmarks in real time, covering areas such as eyes, eyebrows, mouth, nose, and facial contours. After detecting faces, the video clip is then preprocessed.

[0043] Optionally, considering that keypoint alignment can improve the accuracy of feature extraction, after detecting the presence of a face in a video frame, keypoint alignment can also be performed, such as aligning the 68 keypoints of the face to standard positions, and then preprocessing the video segment.

[0044] In this embodiment, the face in the video frame can first be divided into multiple Regions of Interest (ROIs), for example, the forehead, cheeks, and periorbital area can be divided into several ROIs. To improve the accuracy of subsequent processing results, the motion coherence and illumination of each ROI can be made consistent during the segmentation process. After dividing into multiple ROIs, bilateral filtering and Retinex processing are performed on the video frames within each ROI.

[0045] Bilateral filtering is a nonlinear filtering technique that combines information from the spatial domain and the value domain for image smoothing. Compared with traditional linear filtering methods (such as Gaussian filtering), bilateral filtering can preserve edge information well while smoothing the image. In other words, bilateral filtering is beneficial for reducing high-frequency noise in video frames.

[0046] Retinex theory, proposed by Edwin H. Land in the 1960s, aims to explain how the human visual system perceives color and brightness. Retinex processing, based on Retinex theory, is used to enhance the contrast and color of video frames and reduce the impact of lighting changes, specifically reducing low-frequency lighting fluctuations.

[0047] S13, determine the physiological parameters of the person to be tested based on the preprocessed video frame sequence.

[0048] Specifically, physiological parameters of the subject can be determined from the preprocessed video frame sequence based on rPPG, such as the subject's heart rate.

[0049] In this embodiment, after acquiring the video segment of the person to be tested, the video frames of the video segment are preprocessed. Since the preprocessing includes eliminating high-frequency noise and low-frequency illumination fluctuations in the video frames, the high-frequency noise and low-frequency illumination fluctuations in the preprocessed video frame sequence will be less than those in the unprocessed video frame sequence. Therefore, when determining the physiological parameters of the person to be tested based on the preprocessed video frame sequence, the interference of high-frequency noise and low-frequency illumination in determining the physiological parameters can be reduced, thereby improving the accuracy of the determined physiological parameters of the person to be tested.

[0050] In some embodiments, S13 above, determining the physiological parameters of the person under test based on the preprocessed video frame sequence, includes: Based on the preprocessed video frame sequence and vibration imaging technology, the physiological parameters of the person being tested were determined.

[0051] Vibraimage, a non-contact psychological and physiological detection technology, assesses psychological and physiological states by analyzing minute physiological vibrations in the head and neck. This technology is based on the vestibular-emotion reflex theory, which posits a correlation between an individual's vestibular balance regulation ability and their emotional responses.

[0052] Specifically, the preprocessed video frame sequence mentioned above was captured by a camera with sufficient resolution and frame rate to capture minute vibrations of the human body, thus providing the possibility for subsequent application of vibration imaging technology to determine physiological parameters.

[0053] In this embodiment, before analyzing the preprocessed video frame sequence using vibration imaging technology, further filtering and denoising processes can be performed on the preprocessed video frames. Then, the filtered and denoised video frame sequence is analyzed to obtain the physiological vibrations of the person under test in these video frame sequences. Finally, based on the correspondence between these physiological vibrations and emotions, the physiological parameters corresponding to the physiological vibrations of the person under test are determined. Of course, in addition to determining physiological parameters, the psychological parameters of the person under test can also be determined.

[0054] Since vibration imaging technology eliminates the need for physical contact when determining physiological parameters of individuals, it avoids the discomfort and infection risks associated with contact measurements. Furthermore, this technology allows for rapid assessment of physiological parameters within a short timeframe (e.g., 30-60 seconds), making it suitable for rapid screening and real-time monitoring.

[0055] In some embodiments, determining the physiological parameters of the subject based on the preprocessed video frame sequence and vibration imaging technology includes: A1. Determine the micro-motion characteristics of the preprocessed video frame sequence based on vibration imaging technology.

[0056] A2. Enhance the micro-motion features of the skin in the preprocessed video frame sequence to obtain the enhanced video frame sequence.

[0057] Among them, the microkinetic characteristics of the skin refer to the subtle movements that occur when the skin is stimulated by external factors or affected by internal physiological activities.

[0058] Specifically, enhancing skin micro-motion features includes amplifying these features and increasing the contrast between them and other features. It should be noted that when enhancing skin micro-motion features, if the face has been pre-divided into Regions of Interest (ROIs), the skin micro-motion features of each ROI in the video frames can be enhanced separately.

[0059] Optionally, A2 above enhances the micro-motion features of the skin in the preprocessed video frame sequence to obtain an enhanced video frame sequence, including: A21. Perform the following steps on the video frames in the preprocessed video frame sequence: perform pyramid decomposition on the video frames to obtain the decomposed image frame sequence corresponding to the video frames; determine the detail information of the video frames based on the decomposed image frame sequence corresponding to the video frames; extract the frequency component corresponding to the heartbeat from the detail information of the video frames to obtain the heartbeat component; amplify the heartbeat component; and reconstruct the video frames based on the amplified heartbeat component and the decomposed image frame sequence corresponding to the video frames to obtain the reconstructed video frames.

[0060] A22. Based on each of the reconstructed video frames described above, an enhanced video frame sequence is obtained.

[0061] Specifically, the Laplacian pyramid decomposition method or the phase-based magnification method can be used to perform pyramid decomposition on video frames. Of course, other methods can also be used, which will not be elaborated here. The following explanation uses the Laplacian pyramid decomposition method to perform pyramid decomposition on video frames as an example: Assuming the preprocessed video frame (or the video frame corresponding to a certain ROI) is ,in," This indicates the spatial dimension information of the video frame, such as the pixel position within the video frame. This indicates the time dimension information of the video frame, such as the frame number or timestamp of the video frame within the video segment.

[0062] Will Constructing a pyramid using a Gaussian kernel g: ; ; ; in, This is level 0 of the pyramid, i.e., the original image. For the first pyramid class, For the first pyramid The ( ) level is generated by convolution (smoothing). +1) level, For Gaussian kernel function, that is, for After decomposing the pyramid, we obtain the following: The corresponding decomposed image frame sequence ( , , Optionally, the Gaussian kernel of the Gaussian kernel function can be 3x3. For the first Laplace component of the layer ( (corresponding detailed information), which includes edge and high-frequency information at this scale.

[0063] right Using a bandpass filter h(t) (passband 0.7–4Hz), extract the frequency components corresponding to the heartbeat: ; ; in, It is a component of the heartbeat, which is transmitted through the first Layer Laplacian details are filtered in the time dimension Received, the It preserves the periodic signals related to heartbeat. The impulse response of the time-domain bandpass filter is designed to cover typical heartbeat frequencies (approximately 0.7–4 Hz, corresponding to 42–240 bpm). This is the amplification factor, used to control the degree of amplitude enhancement, and it is set according to the actual situation. For example, when... When smaller, set A larger value is used to increase the heart rate component, and vice versa. Smaller.

[0064] Reconstruct the magnified details back into each layer: .

[0065] Where N is the highest level obtained after performing pyramid grading on the video frames.

[0066] Due to Multi-scale detail layers are subjected to temporal bandpass filtering and amplification to generate a product containing amplified micro-scale details. Enhanced video frames with motion information Therefore, it can reveal subtle movements and color fluctuations that are difficult to detect with the naked eye, thus increasing the probability of extracting relevant information about the minute vibrations caused by the heartbeat. A3. Determine the physiological parameters of the subjects based on the enhanced video frame sequence.

[0067] The process of determining the physiological parameters of the test subject from the enhanced video frame sequence is similar to that of determining the physiological parameters of the test subject from the preprocessed video frame sequence, and will not be described in detail here.

[0068] In this embodiment, by enhancing the micro-motion features of the skin in the preprocessed video frame sequence, the enhanced video frame sequence is obtained, and then the physiological parameters of the person being tested are determined based on the enhanced video frame sequence. Since the physiological parameters of the person being tested (such as heart rate) are related to the micro-motion features of the skin, the physiological parameters determined by the above enhancement process are more accurate.

[0069] In some embodiments, considering that the influence of the external environment on the face of the test subject is usually different when the test subject is in motion and when the test subject is at rest, for example, when the test subject is in motion, the external environment in which the test subject is located usually changes due to the change in the position of the test subject (while when the test subject is at rest, the external environment usually does not change), and the ambient light fluctuations of different external environments are usually different, and since the probability of the test subject's facial expression changing when the test subject is in motion is higher than that when the test subject is at rest, it is possible to select whether to enhance the micro-motion features of the skin in the preprocessed video frame sequence according to the state of the test subject.

[0070] That is, by enhancing the micro-motion features of the skin in the preprocessed video frame sequence as described above (A2), an enhanced video frame sequence is obtained, including: When the preprocessed video frame sequence indicates that the subject is in motion, the micro-motion features of the skin in the preprocessed video frame sequence are enhanced to obtain an enhanced video frame sequence.

[0071] Conversely, when the preprocessed video frame sequence indicates that the person under test is in a static state, the physiological parameters of the person under test are determined based on the preprocessed video frame sequence.

[0072] Specifically, the movement of the person being tested can be determined by the differences between video frames in a video frame sequence. For example, if the difference between two adjacent video frames in a video frame sequence is large, the person being tested is determined to be in motion; otherwise, the person being tested is determined to be stationary.

[0073] Since the probability of ambient light fluctuations in the video frame sequence of a person being tested being low when the person is stationary, directly determining the physiological parameters of the person from the preprocessed video frame sequence when the person is stationary is beneficial for improving the speed of physiological parameter determination. Conversely, since the probability of ambient light fluctuations in the video frame sequence of a person being tested being high when the person is in motion, enhancing the skin's micro-motion features in the preprocessed video frame sequence before determining the physiological parameters of the person is beneficial for improving the accuracy of physiological parameter determination.

[0074] In some embodiments, A3, determining the physiological parameters of the person under test based on the enhanced video frame sequence, includes: A31. Calculate the mean of the enhanced video frame sequence in different channels.

[0075] A32. Construct the mean function for each channel based on the mean values ​​of each channel.

[0076] A33. Determine the physiological parameters of the subjects based on the above mean functions.

[0077] Specifically, the aforementioned channel is a video frame channel. Optionally, the channel can be a color channel of the video frame, such as the RGB channel of the video frame.

[0078] The following explanation uses the calculation of the mean value of the RGB channels as an example: For any video frame in the enhanced video frame sequence, calculate the mean value of that video frame across the R, G, and B channels. For example, when calculating the mean value of video frame 1 across the R channel, accumulate the pixel values ​​of each pixel in the R channel of video frame 1 to obtain an accumulated value, and then calculate the mean value of video frame 1 across the R channel based on this accumulated value and the number of pixel values ​​in the R channel. Calculating the mean values ​​of video frame 1 across the G and B channels is similar to calculating the mean value of video frame 1 across the R channel, and will not be described in detail here.

[0079] After calculating the mean values ​​of each video frame in the video frame sequence on the R, G, and B channels, the mean function corresponding to the mean value of the R channel, the mean function corresponding to the mean value of the G channel, and the mean function corresponding to the mean value of the B channel are fitted by combining the timestamp of each video frame in the video segment.

[0080] Assuming the mean function corresponding to the mean of the R channel is r(t), the mean function corresponding to the mean of the G channel is g(t), and the mean function corresponding to the mean of the B channel is b(t), then a two-vector projection can be constructed based on the fitted mean functions: ; .

[0081] It should be noted that the weights of each mean function can be set according to the actual situation.

[0082] Based on the constructed two-vector projection, construct the following functions related to the BVP signal. : .

[0083] in," "Indicates that" The norm of .

[0084] right Peak detection was performed to determine the heartbeat time sequence. ( Indicates the first (a number of heartbeat moments), based on the sequence of heartbeat moments It can calculate the physiological parameters of the person being tested.

[0085] It should be noted that the above describes the method for determining the physiological parameters of the test subject based on the enhanced video frame sequence. The process for determining the physiological parameters of the test subject based on the preprocessed video frame sequence is similar, the difference being that one is determined from the enhanced video frame sequence and the other is determined from the preprocessed video frame sequence. This will not be elaborated further here.

[0086] In some embodiments, when the physiological parameters include heart rate and heart rate variability indicators, the above-mentioned A33, determining the physiological parameters of the subject based on each of the above-mentioned mean functions, includes: A331. Determine the heart rate of the subjects based on the above mean functions.

[0087] A332. Calculate the heart rate variability index of the above-mentioned subjects based on the heart rate.

[0088] Specifically, in determining the heartbeat time sequence Then, the heart rate can be calculated using the following formula. : .

[0089] If the heart rate variability index includes the standard deviation of all sinus RR intervals (Standard Deviation) The root mean square of successive differences (RMSSD) and the root mean square of normal-to-normal intervals (SDNN) are used to obtain the heartbeat time. Then, calculate the adjacent RR interval sequence: ; ; .

[0090] In the formula The average RR interval is given by , and M is the number of heartbeats. The above formula can be used to stably calculate the RMSSD in a 10-second video clip.

[0091] In some embodiments, considering that the longer the video clip is, the more information about the test subject it contains, and that more information about the test subject is beneficial to improving the accuracy of the subsequently obtained heart rate variability index, the final heart rate variability index of the test subject can be determined based on the duration of the video clip. In this case, A332 above, calculating the heart rate variability index of the test subject based on the heart rate, includes: When the duration of the video segment exceeds a preset duration threshold, the heart rate variability index of the subject is calculated based on the heart rate. When the duration of the video segment is not greater than the duration threshold, the heart rate variability index to be corrected for the subject is calculated based on the heart rate, the heartbeat of the subject is calculated based on the heart rate, and the heart rate variability index to be corrected is corrected based on the heartbeat to obtain the final heart rate variability index of the subject.

[0092] The aforementioned preset duration threshold can be determined based on the duration of the video segment required to calculate RMSSD. For example, if the duration of the video segment required to calculate RMSSD is 10 seconds, the preset duration threshold can be set to 10 seconds.

[0093] In this embodiment, when the duration of a video segment exceeds a preset duration threshold, it indicates that sufficient information about the test subject can be obtained from the video segment, meaning the accuracy of the heart rate variability index calculated based on the video segment is high. In this case, the heart rate variability index calculated based on the heart rate is used as the final heart rate variability index for the test subject. Conversely, when the duration of a video segment is not greater than the preset duration threshold, it indicates that less information about the test subject can be obtained from the video segment, meaning the accuracy of the heart rate variability index calculated based on the video segment is low. In this case, a heart rate variability index to be corrected is first calculated based on the heart rate, and then the heart rate variability index to be corrected is corrected based on the calculated heart rate of the test subject. The corrected heart rate variability index is used as the final heart rate variability index for the test subject.

[0094] Alternatively, assume the heart rate variability index to be corrected is and , ; .

[0095] Where M is the effective heart rate (5≤M≤30). and This is an empirical coefficient, which can be determined through calibration experiments. The above correction method can reduce the error in determining the heart rate variability index for video clips shorter than a preset duration threshold, such as reducing the error in determining the heart rate variability index for video clips shorter than 10 seconds. Optionally, it can be set... =0.12、 =0.18.

[0096] In some embodiments, considering that the tolerance for erroneous operations is low in certain scenarios (such as nuclear power plants), a timely warning can be issued when a problem is detected in the physiological parameters of the person being tested. That is, after determining the physiological parameters of the person being tested based on the preprocessed video frame sequence in S13 above, the method further includes: Determine whether the physiological parameters of the person being tested match the warning conditions corresponding to the preset warning level; if the physiological parameters of the person being tested match the warning conditions corresponding to the preset warning level, then trigger the warning corresponding to the matching warning conditions.

[0097] The preset warning levels can include multiple levels, with different levels corresponding to different warning conditions.

[0098] In this embodiment, the warning conditions are related to physiological parameters. For example, when the physiological parameters include SDNN, the warning conditions include SDNN; when the physiological parameters include SDNN and RMSSD, the warning conditions include SDNN and RMSSD; when the physiological parameters include heart rate and SDNN, the warning conditions include heart rate and SDNN. Optionally, the physiological parameters included in the above warning conditions can be numerical ranges. For example, the SDNN included in the warning conditions can be the numerical range corresponding to that SDNN.

[0099] After determining the physiological parameters of the person to be tested, the physiological parameters of the person to be tested are compared with the warning conditions corresponding to the warning level. If the two match, it indicates that the physiological parameters of the person to be tested are abnormal. At this time, issuing the corresponding warning is helpful to promptly identify the person to be tested with abnormal health. Since people with abnormal health have a higher probability of making mistakes at work, warning processing is helpful to reduce the probability of safety accidents in nuclear power plants due to human factors.

[0100] In some embodiments, considering that even healthy individuals may experience short-term abnormalities in physiological parameters, to reduce the probability of false alarms, an alarm is issued only after the abnormality in physiological parameters is detected and persists for a certain period of time. In this case, if the physiological parameters of the person being tested match the alarm conditions corresponding to a preset alarm level, an alarm corresponding to the matching alarm conditions is triggered, including: If the physiological parameters of the person being tested match the physiological parameters corresponding to the preset warning level, and the duration matches the duration corresponding to the preset warning level, then a warning corresponding to the matching warning conditions is triggered. The warning conditions corresponding to the preset warning level include physiological parameters and duration.

[0101] Specifically, the duration is equal to the duration of the video segment. For example, if the duration of the acquired video segment is 10 seconds, the physiological parameters calculated based on the 10-second video segment will be matched with the physiological parameters corresponding to the preset warning level, and the 10 seconds will be matched with the duration corresponding to the preset warning level.

[0102] In this embodiment of the application, considering that the duration of abnormal physiological parameters in a healthy human body is usually short, before triggering the warning, it is not only necessary to determine whether the physiological parameters are abnormal, but also whether the duration of the abnormality is long enough, which is beneficial to improving the accuracy of the warning.

[0103] In some embodiments, the warning conditions can be determined in the following manner, that is, before determining whether the physiological parameters of the person to be tested match the warning conditions corresponding to the preset warning level, the method further includes: B1. Obtain the current ambient light intensity and / or the job type of the person being tested.

[0104] B2. Determine the above-mentioned early warning conditions based on the above-mentioned ambient light intensity and / or the above-mentioned job type.

[0105] The ambient light intensity can be obtained by measuring a photometer. At this time, the electronic device can obtain the ambient light intensity measured by the photometer by establishing a communication connection with the photometer.

[0106] The job type of the test subject can be determined by performing facial recognition on the test subject, and then determining the job type of the test subject based on the facial recognition results and the preset mapping relationship between the face and the job type. Of course, the job type of the test subject can also be determined by other methods, which will not be elaborated here.

[0107] Optionally, considering that the human sympathetic nervous system is excited when ambient light intensity increases, the relationship between the ambient light intensity and the corresponding value or range of the warning condition can be set to be directly proportional. That is, the greater the ambient light intensity, the greater the value (or the upper limit of the range) of the warning condition. Since the heart rate also increases when the human sympathetic nervous system is excited, setting the relationship between the ambient light intensity and the corresponding value or range of the warning condition is beneficial to improving the accuracy of the set warning conditions.

[0108] Optionally, the numerical value or range of the warning conditions can be set according to the safety level of the job type. For example, the higher the safety level of the job type, the smaller the numerical value (or the upper limit of the numerical range) corresponding to the warning conditions. For the same warning level, the higher the safety level of the job type, the smaller the numerical value corresponding to its warning conditions. A higher safety level for the job type indicates a greater relevance of the corresponding work content to the radiation zone. By setting the numerical value or range of the warning conditions according to the safety level of the job type, the warning conditions are more closely matched to the job type, thereby improving the accuracy of the warning conditions.

[0109] In some embodiments, the preset warning levels include Level 1, Level 2, and Level 3 warnings, with Level 3 being the highest. Following the warning corresponding to the triggered and matched warning conditions, the system further includes: If the warning level is higher than the first-level warning mentioned above, then the video clips of the person to be tested are reacquired, and the physiological parameters of the person to be tested are determined based on the reacquired video clips.

[0110] The warning conditions for Level 1 alerts are: heart rate > 100 bpm and SDNN < 40 ms, with a duration of 10 seconds or more. The warning conditions for Level 2 and Level 3 alerts are stricter than those for Level 1 alerts.

[0111] Specifically, video clips of the person under test that trigger a level 2 or level 3 alert can be acquired using face tracking technology. The same processing method as steps S11-S13 above is used to determine the physiological parameters of the person under test in the newly acquired video clips. Optionally, after determining the new physiological parameters, it can be determined whether an alert is still needed for the person under test. If an alert is still needed, the corresponding alert is triggered, and an alert message is sent to a designated terminal. This alert message may include the person under test's identity information and physiological parameters.

[0112] Since the warning level is higher than the first-level warning, and the higher the warning level, the worse the health of the person being tested, re-acquiring the video clips of the person being tested and re-determining their physiological parameters—that is, verifying whether the physiological parameters of the person being tested require a warning through the above processing—is beneficial to improving the accuracy of the determined physiological parameters and the accuracy of the warning.

[0113] In some embodiments, the above-mentioned reacquisition of the video clips of the person being tested includes: The video clips of the person under test are reacquired using a target frame rate, wherein the target frame rate is higher than the frame rate corresponding to the first acquisition of the video clips of the person under test.

[0114] The target frame rate can be 120 frames per second (fps), and the frame rate corresponding to the first video clip of the person being tested can be 30 fps.

[0115] In this embodiment of the application, since a higher frame rate is used to collect video segments of the person being tested, it is beneficial to obtain a sufficient number of video segments more quickly, thereby improving the speed of subsequently determining new physiological parameters and the speed of determining whether an early warning is needed.

[0116] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0117] Corresponding to the health monitoring method for personnel described in the above embodiments, Figure 2 A structural block diagram of a health monitoring device for personnel provided in an embodiment of this application is shown. For ease of explanation, only the parts related to the embodiment of this application are shown.

[0118] Reference Figure 2 The health monitoring device 2 for this person is applied to electronic equipment and includes: a video segment acquisition module 21, a preprocessing module 22, and a physiological parameter determination module 23. Among them: The video segment acquisition module 21 is used to acquire video segments of the person to be tested, wherein the video segments include at least two video frames.

[0119] The aforementioned video clips can be video clips acquired during the movement of the person being tested, or video clips acquired when the person being tested is stationary.

[0120] The number of people to be tested can be equal to 1 or greater than 1.

[0121] The preprocessing module 22 is used to preprocess the video frames of the above video segment to obtain a preprocessed video frame sequence, wherein the preprocessing includes eliminating high-frequency noise and low-frequency illumination fluctuations of the above video frames.

[0122] Optionally, considering that keypoint alignment can improve the accuracy of feature extraction, after detecting the presence of a face in a video frame, keypoint alignment can also be performed, such as aligning the 68 keypoints of the face to standard positions, and then preprocessing the video segment.

[0123] The physiological parameter determination module 23 is used to determine the physiological parameters of the person to be tested based on the preprocessed video frame sequence.

[0124] In this embodiment, after acquiring the video segment of the person to be tested, the video frames of the video segment are preprocessed. Since the preprocessing includes eliminating high-frequency noise and low-frequency illumination fluctuations in the video frames, the high-frequency noise and low-frequency illumination fluctuations in the preprocessed video frame sequence will be less than those in the unprocessed video frame sequence. Therefore, when determining the physiological parameters of the person to be tested based on the preprocessed video frame sequence, the interference of high-frequency noise and low-frequency illumination in determining the physiological parameters can be reduced, thereby improving the accuracy of the determined physiological parameters of the person to be tested.

[0125] Optionally, the physiological parameter determination module 23 includes: The vibration measurement unit is used to determine the physiological parameters of the person being tested based on the preprocessed video frame sequence and vibration imaging technology.

[0126] Optionally, the vibration measurement unit mentioned above includes: The micro-motion feature determination unit is used to determine the micro-motion features of the preprocessed video frame sequence based on vibration imaging technology.

[0127] The micro-motion feature enhancement unit is used to enhance the micro-motion features of the skin in the preprocessed video frame sequence to obtain the enhanced video frame sequence.

[0128] The enhanced video frame sequence determination unit is used to determine the physiological parameters of the person under test based on the enhanced video frame sequence.

[0129] Optionally, considering that the influence of the external environment on the face of the test subject is usually different when the test subject is in motion and when the test subject is at rest, the above-mentioned micro-motion feature enhancement unit is specifically used for: When the preprocessed video frame sequence indicates that the subject is in motion, the micro-motion features of the skin in the preprocessed video frame sequence are enhanced to obtain an enhanced video frame sequence.

[0130] Optionally, the physiological parameter determination module 23 includes: When the preprocessed video frame sequence indicates that the person under test is in a static state, the physiological parameters of the person under test are determined based on the preprocessed video frame sequence.

[0131] Optionally, the aforementioned micro-motion feature enhancement unit is specifically used for: The following steps are performed on the video frames in the preprocessed video frame sequence: the video frames are decomposed into pyramids to obtain the decomposed image frame sequence corresponding to the video frames; the detailed information of the video frames is determined based on the decomposed image frame sequence corresponding to the video frames; the frequency components corresponding to the heartbeats in the detailed information of the video frames are extracted to obtain the heartbeat components; the heartbeat components are amplified; and the video frames are reconstructed based on the amplified heartbeat components and the decomposed image frame sequence corresponding to the video frames to obtain the reconstructed video frames. The enhanced video frame sequence is obtained based on each of the reconstructed video frames described above.

[0132] Specifically, the Laplacian pyramid decomposition method or the Phase-based Magnification method can be used to decompose video frames into pyramids. Of course, other methods can also be used to decompose video frames into pyramids, which will not be elaborated here.

[0133] Optionally, the enhanced video frame sequence determination unit described above is specifically used for: Calculate the mean of the enhanced video frame sequence across different channels; Construct a mean function for each channel based on the mean value of each channel; The physiological parameters of the subjects were determined based on the aforementioned mean functions.

[0134] Optionally, when the physiological parameters include heart rate and heart rate variability indicators, the above-mentioned determination of the physiological parameters of the subject based on the respective mean functions includes: The heart rate of the subjects was determined based on the aforementioned mean functions. The heart rate variability index of the subjects was calculated based on the above heart rate.

[0135] Specifically, in determining the heartbeat time sequence Then, the heart rate can be calculated using the following formula. : .

[0136] If the heart rate variability index includes the standard deviation of all sinus RR intervals (Standard Deviation) The root mean square of successive differences (RMSSD) and the root mean square of normal-to-normal intervals (SDNN) are used to obtain the heartbeat time. Then, calculate the adjacent RR interval sequence: ; ; .

[0137] In the formula The average RR interval is given by , and M is the number of heartbeats. The above formula can be used to stably calculate the RMSSD in a 10-second video clip.

[0138] Optionally, considering that a longer video clip contains more information about the test subject, and more information about the test subject is beneficial to improving the accuracy of the subsequent heart rate variability index, the final heart rate variability index of the test subject can be determined based on the length of the video clip. In this case, the above-mentioned calculation of the heart rate variability index of the test subject based on the aforementioned heart rate includes: When the duration of the aforementioned video segment exceeds a preset duration threshold, the heart rate variability index of the aforementioned test subject is calculated based on the aforementioned heart rate. When the duration of the aforementioned video segment is not greater than the aforementioned duration threshold, the heart rate variability index to be corrected for the aforementioned person under test is calculated based on the aforementioned heart rate, the heartbeat of the aforementioned person under test is calculated based on the aforementioned heart rate, and the heart rate variability index to be corrected is corrected based on the aforementioned heartbeat to obtain the final heart rate variability index of the aforementioned person under test.

[0139] The aforementioned preset duration threshold can be determined based on the duration of the video segment required to calculate RMSSD. For example, if the duration of the video segment required to calculate RMSSD is 10 seconds, the preset duration threshold can be set to 10 seconds.

[0140] In this embodiment, when the duration of a video segment exceeds a preset duration threshold, it indicates that sufficient information about the test subject can be obtained from the video segment, meaning the accuracy of the heart rate variability index calculated based on the video segment is high. In this case, the heart rate variability index calculated based on the heart rate is used as the final heart rate variability index for the test subject. Conversely, when the duration of a video segment is not greater than the preset duration threshold, it indicates that less information about the test subject can be obtained from the video segment, meaning the accuracy of the heart rate variability index calculated based on the video segment is low. In this case, a heart rate variability index to be corrected is first calculated based on the heart rate, and then the heart rate variability index to be corrected is corrected based on the calculated heart rate of the test subject. The corrected heart rate variability index is used as the final heart rate variability index for the test subject.

[0141] Alternatively, assume the heart rate variability index to be corrected is and , ; .

[0142] Where M is the effective heart rate (5≤M≤30). and This is an empirical coefficient, which can be determined through calibration experiments. The above correction method can reduce the error in determining the heart rate variability index for video clips shorter than a preset duration threshold, such as reducing the error in determining the heart rate variability index for video clips shorter than 10 seconds. Optionally, it can be set... =0.12、 =0.18.

[0143] Optionally, considering the low tolerance for operational errors in certain scenarios (such as nuclear power plants), a timely warning can be issued when a problem is detected in the physiological parameters of the person being tested. That is, the health monitoring device 2 for this person also includes: The early warning condition matching module is used to determine whether the physiological parameters of the person to be tested match the early warning conditions corresponding to the preset early warning level after the physiological parameters of the person to be tested are determined based on the preprocessed video frame sequence.

[0144] The early warning triggering module is used to trigger an early warning corresponding to the matching early warning conditions if the physiological parameters of the person to be tested match the early warning conditions corresponding to the preset early warning level.

[0145] The preset warning levels can include multiple levels, with different levels corresponding to different warning conditions.

[0146] In this embodiment, the warning conditions are related to physiological parameters. For example, when the physiological parameters include SDNN, the warning conditions include SDNN; when the physiological parameters include SDNN and RMSSD, the warning conditions include SDNN and RMSSD; when the physiological parameters include heart rate and SDNN, the warning conditions include heart rate and SDNN. Optionally, the physiological parameters included in the above warning conditions can be numerical ranges. For example, the SDNN included in the warning conditions can be the numerical range corresponding to that SDNN.

[0147] After determining the physiological parameters of the person to be tested, the physiological parameters of the person to be tested are compared with the warning conditions corresponding to the warning level. If the two match, it indicates that the physiological parameters of the person to be tested are abnormal. At this time, issuing the corresponding warning is helpful to promptly identify the person to be tested with abnormal health conditions. Since people with abnormal health conditions have a higher probability of making mistakes at work, warning processing is helpful to reduce the probability of safety accidents occurring in nuclear power plants.

[0148] Optionally, considering that even healthy individuals may experience short-term abnormalities in physiological parameters, to reduce the probability of false alarms, an alarm is only issued after an abnormality in physiological parameters is detected and persists for a certain period of time. In this case, the aforementioned alarm triggering module is specifically used for: If the physiological parameters of the person being tested match the physiological parameters corresponding to the preset warning level, and the duration matches the duration corresponding to the preset warning level, then the warning corresponding to the matching warning conditions is triggered. The warning conditions corresponding to the preset warning level include physiological parameters and duration.

[0149] Specifically, the duration is equal to the duration of the video segment. For example, if the duration of the acquired video segment is 10 seconds, the physiological parameters calculated based on the 10-second video segment will be matched with the physiological parameters corresponding to the preset warning level, and the 10 seconds will be matched with the duration corresponding to the preset warning level.

[0150] Optionally, the health monitoring device 2 for the person also includes: The ambient light intensity acquisition module is used to acquire the current ambient light intensity and / or the job type of the person being tested before determining whether the physiological parameters of the person being tested match the warning conditions corresponding to the preset warning level.

[0151] The warning condition determination module is used to determine the warning conditions based on the ambient light intensity and / or the job type.

[0152] The ambient light intensity can be obtained by measuring a photometer. At this time, the electronic device can obtain the ambient light intensity measured by the photometer by establishing a communication connection with the photometer.

[0153] The job type of the test subject can be determined by performing facial recognition on the test subject, and then determining the job type of the test subject based on the facial recognition results and the preset mapping relationship between the face and the job type. Of course, the job type of the test subject can also be determined by other methods, which will not be elaborated here.

[0154] Optionally, considering that the human sympathetic nervous system is excited when ambient light intensity increases, the relationship between the ambient light intensity and the corresponding value or range of the warning condition can be set to be directly proportional. That is, the greater the ambient light intensity, the greater the value (or the upper limit of the range) of the warning condition. Since the heart rate also increases when the human sympathetic nervous system is excited, setting the relationship between the ambient light intensity and the corresponding value or range of the warning condition is beneficial to improving the accuracy of the set warning conditions.

[0155] Optionally, the corresponding values ​​or ranges of warning conditions can be set according to the safety level of the job type. For example, the higher the safety level of the job type, the smaller the value (or the upper limit of the value range) corresponding to the warning condition. For the same warning level, the higher the safety level of the job type, the smaller the value corresponding to its warning condition. A higher safety level for the job type indicates a greater correlation between the corresponding work content and the radiation zone.

[0156] Optionally, the preset warning levels include Level 1, Level 2, and Level 3 warnings, with Level 3 being the highest. The health monitoring device 2 for the person also includes: The verification module is used to reacquire the video clips of the person under test after the warning corresponding to the above-mentioned triggering and matching warning conditions. If the warning level is higher than the above-mentioned first-level warning, the module will reacquire the video clips of the person under test and determine the physiological parameters of the person under test based on the reacquired video clips.

[0157] Optionally, the above-mentioned reacquisition of video clips of the aforementioned test subjects includes: The video clips of the person under test are reacquired using a target frame rate, wherein the target frame rate is higher than the frame rate corresponding to the first acquisition of the video clips of the person under test.

[0158] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0159] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 3 As shown, the electronic device 3 of this embodiment includes: at least one processor 30 ( Figure 3 The diagram shows only one processor, a memory 31, and a computer program 32 stored in the memory 31 and executable on the at least one processor 30, wherein the processor 30 executes the computer program 32 to implement the steps in any of the above method embodiments.

[0160] The electronic device 3 can be a desktop computer, laptop, handheld computer, or cloud server, etc. This electronic device may include, but is not limited to, a processor 30 and a memory 31. Those skilled in the art will understand that... Figure 3 This is merely an example of electronic device 3 and does not constitute a limitation on electronic device 3. It may include more or fewer components than shown in the figure, or combine certain components, or different components. For example, it may also include input / output devices, network access devices, etc.

[0161] The processor 30 may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0162] In some embodiments, the memory 31 may be an internal storage unit of the electronic device 3, such as a hard disk or memory of the electronic device 3. In other embodiments, the memory 31 may be an external storage device of the electronic device 3, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the electronic device 3. Furthermore, the memory 31 may include both internal and external storage units of the electronic device 3. The memory 31 is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory 31 can also be used to temporarily store data that has been output or will be output.

[0163] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0164] This application also provides a network device, which includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor executes the computer program to implement the steps in any of the above method embodiments.

[0165] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps in the above-described method embodiments.

[0166] This application provides a computer program product that, when run on an electronic device, enables the electronic device to implement the steps described in the various method embodiments above.

[0167] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a photographic device / electronic device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0168] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0169] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0170] In the embodiments provided in this application, it should be understood that the disclosed apparatus / network devices and methods can be implemented in other ways. For example, the apparatus / network device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0171] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0172] It should be noted that the information collection process (such as the facial image collection process) / feature extraction process involved in this application is carried out with the user's knowledge and permission. That is, the information collection process / feature extraction process complies with the requirements of laws and regulations and does not constitute an act that harms the public interest.

[0173] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for health monitoring of personnel, characterized in that, include: Acquire a video clip of the person to be tested, wherein the video clip includes at least two video frames; The video frames of the video segment are preprocessed to obtain a preprocessed video frame sequence, wherein the preprocessing includes eliminating high-frequency noise and low-frequency illumination fluctuations in the video frames; The physiological parameters of the person being tested are determined based on the preprocessed video frame sequence.

2. The method for health monitoring of personnel as described in claim 1, characterized in that, The step of determining the physiological parameters of the person under test based on the preprocessed video frame sequence includes: The physiological parameters of the person under test are determined based on the preprocessed video frame sequence and vibration imaging technology.

3. The method for health monitoring of personnel as described in claim 2, characterized in that, The step of determining the physiological parameters of the person under test based on the preprocessed video frame sequence and vibration imaging technology includes: The micro-motion characteristics of the preprocessed video frame sequence are determined based on vibration imaging technology; Enhance the micro-motion features of the skin in the preprocessed video frame sequence to obtain an enhanced video frame sequence; The physiological parameters of the person being tested are determined based on the enhanced video frame sequence.

4. The method for health monitoring of personnel as described in claim 3, characterized in that, The process of enhancing the micro-motion features of the skin in the preprocessed video frame sequence to obtain the enhanced video frame sequence includes: When the preprocessed video frame sequence indicates that the person under test is in motion, the micro-motion features of the skin in the preprocessed video frame sequence are enhanced to obtain an enhanced video frame sequence.

5. The method for health monitoring of personnel as described in claim 4, characterized in that, Also includes: When the preprocessed video frame sequence indicates that the person under test is in a static state, the physiological parameters of the person under test are determined based on the preprocessed video frame sequence.

6. The method for health monitoring of personnel as described in claim 3, characterized in that, The process of enhancing the micro-motion features of the skin in the preprocessed video frame sequence to obtain the enhanced video frame sequence includes: The following steps are performed on the video frames in the preprocessed video frame sequence: the video frame is decomposed into a pyramid to obtain the decomposed image frame sequence corresponding to the video frame; the detailed information of the video frame is determined based on the decomposed image frame sequence corresponding to the video frame; the frequency component corresponding to the heartbeat in the detailed information of the video frame is extracted to obtain the heartbeat component; the heartbeat component is amplified; and the video frame is reconstructed based on the amplified heartbeat component and the decomposed image frame sequence corresponding to the video frame to obtain the reconstructed video frame. An enhanced video frame sequence is obtained based on each of the reconstructed video frames.

7. The method for health monitoring of personnel as described in claim 3, characterized in that, The step of determining the physiological parameters of the person under test based on the enhanced video frame sequence includes: Calculate the mean of the enhanced video frame sequence across different channels; Construct a mean function for each channel based on the mean value of each channel; The physiological parameters of the test subject are determined based on each of the aforementioned mean functions.

8. The method for health monitoring of personnel as described in claim 7, characterized in that, The step of determining the physiological parameters of the subject based on each of the mean functions includes: The heart rate of the person being tested is determined based on each of the aforementioned mean functions; The heart rate variability index of the subject is calculated based on the heart rate.

9. The method for health monitoring of personnel as described in claim 8, characterized in that, The calculation of the heart rate variability index of the subject based on the heart rate includes: When the duration of the video segment exceeds a preset duration threshold, the heart rate variability index of the person being tested is calculated based on the heart rate. When the duration of the video segment is not greater than the duration threshold, the heart rate variability index to be corrected for the person under test is calculated based on the heart rate, the heartbeat of the person under test is calculated based on the heart rate, and the heart rate variability index to be corrected is corrected based on the heartbeat to obtain the final heart rate variability index of the person under test.

10. The method for health detection of personnel as described in any one of claims 1 to 9, characterized in that, After determining the physiological parameters of the subject based on the preprocessed video frame sequence, the method further includes: Determine whether the physiological parameters of the person being tested match the warning conditions corresponding to the preset warning level; If the physiological parameters of the person being tested match the warning conditions corresponding to the preset warning level, then the warning corresponding to the matching warning conditions is triggered.

11. The method for health detection of personnel as described in claim 10, characterized in that, If the physiological parameters of the person being tested match the warning conditions corresponding to the preset warning level, then the warning corresponding to the matching warning conditions is triggered, including: If the physiological parameters of the person being tested match the physiological parameters corresponding to the preset warning level, and the duration matches the duration corresponding to the preset warning level, then a warning corresponding to the matching warning conditions is triggered, wherein the warning conditions corresponding to the preset warning level include physiological parameters and duration.

12. The method for health detection of personnel as described in claim 10, characterized in that, Before determining whether the physiological parameters of the person being tested match the pre-set warning conditions for the warning level, the method further includes: Obtain the current ambient light intensity and / or the job type of the person being tested; The warning conditions are determined based on the ambient light intensity and / or the job type.

13. The method for health detection of personnel as described in claim 10, characterized in that, The preset warning levels include Level 1, Level 2, and Level 3 warnings, with Level 3 being the highest. Following the warning corresponding to the triggered and matched warning conditions, the following is also included: If the warning level is higher than the first-level warning, the video clip of the person to be tested is reacquired, and the physiological parameters of the person to be tested are determined based on the reacquired video clip.

14. The method for health detection of personnel as described in claim 13, characterized in that, The process of reacquiring the video clips of the person being tested includes: The video segment of the person under test is reacquired using a target frame rate, wherein the target frame rate is higher than the frame rate corresponding to the first acquisition of the video segment of the person under test.

15. A health monitoring device for personnel, characterized in that, include: The video clip acquisition module is used to acquire video clips of the person to be tested, wherein the video clips include at least two video frames; A preprocessing module is used to preprocess the video frames of the video segment to obtain a preprocessed video frame sequence, wherein the preprocessing includes eliminating high-frequency noise and low-frequency illumination fluctuations in the video frames; The physiological parameter determination module is used to determine the physiological parameters of the person to be tested based on the preprocessed video frame sequence.

16. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1 to 14.

17. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 14.

18. A computer program product, characterized in that, Includes a computer program, which, when run, causes the method as described in any one of claims 1 to 14 to be performed.

Images