Information determination method and electronic device

Abstract

Embodiments of the present application disclose an information determination method, which comprises: collecting, by a first image acquisition component, a first to-be-processed image of a first skin region of a target object, and collecting, by a second image acquisition component, a second to-be-processed image of a second skin region of the target object; generating a first heart rate signal based on a parameter of a first pixel point in the first to-be-processed image and a first time corresponding to the collection of the first to-be-processed image; generating a second heart rate signal based on a parameter of a second pixel point in the second to-be-processed image and a second time corresponding to the collection of the second to-be-processed image; determining blood vessel volume change information and time difference information of pulse wave transmission based on the first heart rate signal and the second heart rate signal; and processing the blood vessel volume change information and the time difference information by a blood pressure detection model to obtain target blood pressure information of the target object. Embodiments of the present application also disclose an electronic device.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of information processing, in particular to an information determination method and an electronic device. BACKGROUND

[0002] Blood pressure information is one of the important human physiological signs, which can reflect many important human health information, and regular blood pressure information measurement can effectively prevent physiological diseases. In related technologies, blood pressure information is measured by inflating an arm cuff worn on the user's arm to pressurize the arm, and when the arm pressure reaches a certain pressure, the arm artery blood pressure stops flowing, then the cuff pressure is reduced, and when it is reduced to a certain pressure, the upper arm dynamic blood pressure starts to flow to detect the blood pressure information. However, this measurement method requires the user to wear a cuff, and the measurement process is complex and time-consuming. SUMMARY

[0003] To solve the above technical problems, the embodiments of the present application expect to provide an information determination method and an electronic device, and the technical scheme of the present application is as follows:

[0004] An information determination method, the method comprising:

[0005] acquiring, by a first image acquisition component, a first to-be-processed image of a first skin region of a target object, and acquiring, by a second image acquisition component, a second to-be-processed image of a second skin region of the target object; wherein the first skin region corresponds to a different object part from the second skin region;

[0006] generating a first heart rate signal based on a parameter of a first pixel point in the first to-be-processed image and a first time corresponding to acquisition of the first to-be-processed image;

[0007] generating a second heart rate signal based on a parameter of a second pixel point in the second to-be-processed image and a second time corresponding to acquisition of the second to-be-processed image;

[0008] determining blood vessel volume change information and time difference information of pulse wave transmission based on the first heart rate signal and the second heart rate signal;

[0009] processing the blood vessel volume change information and the time difference information of pulse wave transmission by a blood pressure detection model to obtain target blood pressure information of the target object.

[0010] In the above scheme, the determination of the blood vessel volume change information and the time difference information of pulse wave transmission based on the first heart rate signal and the second heart rate signal comprises:

[0011] The first heart rate signal and the second heart rate signal are respectively adjusted to obtain a first target signal and a second target signal;

[0012] The time difference information is determined based on a third time corresponding to a first target sample in the first target signal and a fourth time corresponding to a second target sample in the second target signal.

[0013] A first waveform corresponding to the first target signal and a second waveform corresponding to the second target signal are determined, and the blood vessel volume variation information is determined based on the first waveform and the second waveform.

[0014] In the above scheme, the first heart rate signal and the second heart rate signal are respectively adjusted to obtain a first target signal and a second target signal, comprising:

[0015] The first heart rate signal is subjected to interpolation processing based on the first time and a target time interval to obtain a first interpolation signal.

[0016] The second heart rate signal is subjected to interpolation processing based on the second time and the target time interval to obtain a second interpolation signal.

[0017] The first interpolation signal and the second interpolation signal are respectively adjusted based on a fifth time corresponding to a peak value in the first interpolation signal and a sixth time corresponding to a valley value, and a seventh time corresponding to a peak value in the second interpolation signal and an eighth time corresponding to a valley value, to obtain the first target signal and the second target signal.

[0018] In the above scheme, the time difference information is determined based on a third time corresponding to a first target sample in the first target signal and a fourth time corresponding to a second target sample in the second target signal, comprising:

[0019] A first time difference between a first sub-time corresponding to a peak value in the first target signal and a third sub-time corresponding to a peak value in the second target signal is calculated.

[0020] A second time difference between a second sub-time corresponding to a valley value in the first target signal and a fourth sub-time corresponding to a valley value in the second target signal is calculated, wherein the peak value in the first target signal has a corresponding relationship with the peak value in the second target signal, the valley value in the first target signal has a corresponding relationship with the valley value in the second target signal, the third time includes the first sub-time and the second sub-time, and the fourth time includes the third sub-time and the fourth sub-time.

[0021] The plurality of first time differences and the plurality of second time differences are operated to determine the time difference information.

[0022] In the scheme, the determination of the first waveform corresponding to the first target signal and the second waveform corresponding to the second target signal comprises:

[0023] Segmenting the first target signal based on adjacent peak values in the first target signal to obtain a plurality of first sub-signals, and segmenting the second target signal based on adjacent peak values in the second target signal to obtain a plurality of second sub-signals;

[0024] Determining the waveform of each first sub-signal to obtain the first waveform;

[0025] Determining the waveform of each second sub-signal to obtain the second waveform.

[0026] In the scheme, the first heart rate signal is generated based on the parameters of the first pixel points in the first to-be-processed image and the first time when the first to-be-processed image is collected, comprising:

[0027] For each frame of the first to-be-processed image, the first pixel value of the pixel point at the first region in the first to-be-processed image is determined; wherein the first to-be-processed image is collected under the condition that the first image collection component is in contact with the first skin region;

[0028] The value of the target color channel is obtained from the first pixel value, and the first reference value of the first to-be-processed image is obtained by operating the value of the target color channel;

[0029] The first heart rate signal is generated based on the first reference value and the first time.

[0030] In the scheme, the second heart rate signal is generated based on the parameters of the second pixel points in the second to-be-processed image and the second time when the second to-be-processed image is collected, comprising:

[0031] The face region of the target object in each frame of the second to-be-processed image is regionally divided to obtain a plurality of different second regions; wherein the second skin region comprises the face region; wherein the second to-be-processed image is collected under the condition that the second image collection component is not in contact with the second skin region;

[0032] The reference signal corresponding to each type of second region is determined based on the second pixel value of each type of second region and the second time;

[0033] Each reference signal is processed to obtain the second heart rate signal;

[0034] In the scheme, the reference signal corresponding to each type of second region is determined based on the second pixel value of each type of second region and the second time, comprising:

[0035] determine a second reference value of each second region based on a plurality of second pixel values of the second region;

[0036] obtain a reference signal corresponding to each type of second region based on a plurality of second reference values of the second region and the second time.

[0037] In the above scheme, the processing of each reference signal to obtain the second heart rate signal comprises:

[0038] filtering processing is performed on each reference signal, and each processed reference signal is synthesized in a color channel to obtain a target value;

[0039] determine a candidate signal based on a plurality of target values and the second time;

[0040] decompose the candidate signal, and determine a signal of a target frequency band from the decomposed signal to obtain the second heart rate signal.

[0041] In the above scheme, the processing of the blood vessel volume change information and the time difference information by the blood pressure detection model to obtain the target blood pressure information of the target object comprises:

[0042] obtain attribute information of the target object;

[0043] process the blood vessel volume change information, the time difference information and the attribute information by the blood pressure detection model to obtain target initial blood pressure information of the target object;

[0044] obtain a correlation relationship between the blood pressure information and the initial blood pressure information of the target object, and determine the target blood pressure information of the target object based on the target initial blood pressure information and the correlation relationship.

[0045] An electronic device, comprising: a first image acquisition component, a second image acquisition component and a processor;

[0046] The first image acquisition component is configured to acquire a first to-be-processed image of a first skin region of a target object;

[0047] The second image acquisition component is configured to acquire a second to-be-processed image of a second skin region of the target object; wherein the object part corresponding to the first skin region and the object part corresponding to the second skin region are different;

[0048] The processor is configured to generate a first heart rate signal based on a parameter of a first pixel point in the first to-be-processed image and a first time corresponding to the acquisition of the first to-be-processed image.

[0049] generate a second heart rate signal based on parameters of second pixel points in the second to-be-processed image and a second time corresponding to acquisition of the second to-be-processed image;

[0050] determine blood vessel volume variation information and time difference information of pulse wave transmission based on the first heart rate signal and the second heart rate signal;

[0051] process, by a blood pressure detection model, the blood vessel volume variation information and the time difference information to obtain target blood pressure information of the target object.

[0052] The information determination method and the electronic device provided in the embodiments of the present application determine blood vessel volume variation information and time difference information of pulse wave transmission based on a first to-be-processed image of a first skin region of a target object acquired by a first image acquisition component and a second to-be-processed image of a second skin region of the target object acquired by a second image acquisition component, wherein the object part corresponding to the first skin region and the object part corresponding to the second skin region are different. A first heart rate signal is generated based on parameters of first pixel points in the first to-be-processed image and a first time corresponding to acquisition of the first to-be-processed image. A second heart rate signal is generated based on parameters of second pixel points in the second to-be-processed image and a second time corresponding to acquisition of the second to-be-processed image. Blood vessel volume variation information and time difference information of pulse wave transmission are determined based on the first heart rate signal and the second heart rate signal. The blood vessel volume variation information and the time difference information of pulse wave transmission are processed by a blood pressure detection model to obtain target blood pressure information of the target object. In this way, only the first to-be-processed image and the second to-be-processed image of two different skin regions of the target object, the first time corresponding to acquisition of the first to-be-processed image, and the second time corresponding to acquisition of the second to-be-processed image are needed to determine the blood vessel volume variation information and the time difference information of pulse wave transmission of the target object. Then, the blood vessel volume variation information and the time difference information of pulse wave transmission are processed by the blood pressure detection model to obtain the target blood pressure information of the target object. The complexity of the measurement process of measuring the target blood pressure information is reduced, the measurement time is saved, and the problem of complex process and large time consumption in related art for measuring blood pressure information is solved. BRIEF DESCRIPTION OF DRAWINGS

[0053] Figure 1 A flowchart of an information determination method provided in the embodiments of the present application is shown.

[0054] Figure 2 A flowchart of another information determination method provided in the embodiments of the present application is shown.

[0055] Figure 3A scene schematic diagram for collecting the first to-be-processed image and collecting the second to-be-processed image in another information determination method provided in the embodiments of the present application is shown in FIG. 1.

[0056] Figure 4 A structural schematic diagram of an electronic device provided in the embodiments of the present application is shown in FIG. 2. DETAILED DESCRIPTION

[0057] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application.

[0058] It should be understood that the specific embodiments described herein are merely intended to explain the present application, and are not intended to limit the present application.

[0059] The embodiments of the present application provide an information determination method, which can be applied to an electronic device. As shown in FIG. 1, the method comprises the following steps. Figure 1

[0060] In step 101, a first to-be-processed image of a first skin region of a target object is collected by a first image collection component, and a second to-be-processed image of a second skin region of the target object is collected by a second image collection component.

[0061] The object part corresponding to the first skin region and the object part corresponding to the second skin region are different. The electronic device can be a device with image collection and processing functions; the first image collection component and the second image collection component can be different image collection components on the electronic device. The number of the first to-be-processed image and the second to-be-processed image is both plural; the target object is an object with vital signs. In a feasible implementation manner, the electronic device can be a mobile phone, and the first image collection component and the second image collection component can be a rear camera and a front camera on the mobile phone respectively.

[0062] In the embodiments of the present application, the first to-be-processed image can be obtained by collecting images of the first skin region of the target object at different time points by the first image collection component; the second to-be-processed image can be obtained by collecting images of the second skin region of the target object at different time points by the second image collection component. The collection manner of the first image collection component for collecting the first to-be-processed image can be the same as or different from the collection manner of the second image collection component for collecting the second to-be-processed image.

[0063] ​Of course, the first image acquisition component can also be used to collect a first to-be-processed video of the first skin area of the target object, and the images in the first to-be-processed video are screened to obtain a first to-be-processed image; and the second image acquisition component can be used to collect a second to-be-processed video of the second skin area of the target object, and the images in the second to-be-processed video are screened to obtain a second to-be-processed image. The acquisition manner of the first image acquisition component for collecting the first to-be-processed video can be the same as or different from the acquisition manner of the second image acquisition component for collecting the second to-be-processed video. The acquisition manner includes a contact acquisition manner and a non-contact acquisition manner.

[0064] In a possible implementation, the target object can be a person, the electronic device can be a mobile phone of the person, the first skin area can be a face area of the person, and the second skin area can be a finger area of the person. The first to-be-processed video of the finger area of the person can be collected in the contact acquisition manner, specifically, the rear camera of the mobile phone can be used to collect the first to-be-processed video of the finger area of the person in a target time period by contacting the finger area of the person. The second to-be-processed video of the face area can be collected in the non-contact acquisition manner, specifically, the front camera of the mobile phone can be used to collect the second to-be-processed video of the face area in the target time period by not contacting the face area and by the face area being within the field of view of the front camera.

[0065] In the embodiments of the present application, the images in the first to-be-processed video and the images in the second to-be-processed video can be filtered to obtain a plurality of first to-be-processed images and a plurality of second to-be-processed images, based on the time of the images in the first to-be-processed video, the time of the images in the second to-be-processed video, and a target time threshold.

[0066] It should be noted that when the first to-be-processed video is collected in the contact acquisition manner and the second to-be-processed video is collected in the non-contact acquisition manner, the images in the first to-be-processed video and the images in the second to-be-processed video can be filtered based on the images in the first to-be-processed video and the target time threshold, to obtain the first to-be-processed images and the second to-be-processed images, so as to improve the accuracy of the blood pressure information of the target object determined subsequently.

[0067] In a feasible implementation, the first to-be-processed video is a video for a finger region, and the second to-be-processed video is a video for a face region; taking the Nth image in the first to-be-processed video as an example, the time of the Nth image in the first to-be-processed video can be acquired, denoted as t1, and the target time threshold is 5 ms, so that the target time range corresponding to the Nth image can be determined as (t1-5 ms, t1+5 ms), and then whether there is a target image in the second to-be-processed video whose time is in the target time range can be determined based on the time of the image in the second to-be-processed video in the second to-be-processed video and the target time range; if there is, the Nth image and the target image are retained; if there is not, the Nth image is deleted; in this way, each image in the first to-be-processed video is taken as a reference, and the images in the first to-be-processed video and the images in the second to-be-processed video are filtered in combination with the determined target time range corresponding to each image, to obtain the first to-be-processed image and the second to-be-processed image; through the target time threshold, it can be ensured that the time interval between the finally determined first to-be-processed image and the second to-be-processed image is less than the preset time interval, so as to realize frame matching between the first to-be-processed image and the second to-be-processed image, and improve the accuracy of the subsequently determined target blood pressure information.

[0068] In step 102, a first heart rate signal is generated based on a parameter of a first pixel point in the first to-be-processed image and a first time corresponding to acquisition of the first to-be-processed image.

[0069] The first time can be a time point at which the first image acquisition component acquires the first to-be-processed image; when the first to-be-processed image is an image in the first to-be-processed video, the first time can be the time of the first to-be-processed image in the first to-be-processed video. The first pixel point is all pixel points in the first to-be-processed image; of course, the first pixel point can also be part of the pixel points in the first to-be-processed image obtained by screening all the pixel points in the first to-be-processed image; the parameter of the first pixel point can be the pixel value of the first pixel point.

[0070] In the embodiments of the present application, for each first to-be-processed image, a first reference value corresponding to the first to-be-processed image can be determined based on the pixel value of the first pixel point in the first to-be-processed image, and then a first heart rate signal can be generated based on the first reference value corresponding to each first to-be-processed image and the first time corresponding to acquisition of each first to-be-processed image. Each first to-be-processed image corresponds to a first reference value.

[0071] In a feasible implementation, the first heart rate signal can be a first photo plethysmography (PPG) signal.

[0072] It should be noted that a first sub-time difference between the first time corresponding to the collection of the first to-be-processed image and the second time corresponding to the collection of the second to-be-processed image satisfies a first target time difference threshold, so as to avoid that the first sub-time difference is too large to cause the first heart rate signal and the second heart rate signal to be not contrastive, and further cause the determined target blood pressure information to be inaccurate; the first target time difference threshold is pre-set. Preferably, the first to-be-processed image and the second to-be-processed image are collected at the same time.

[0073] In step 103, a second heart rate signal is generated based on the parameter of the second pixel point in the second to-be-processed image and the second time corresponding to the collection of the second to-be-processed image.

[0074] The second time can be a time point when the second to-be-processed image is collected by the second image collection component; when the second to-be-processed image is an image in the second to-be-processed video, the second time can be a time of the second to-be-processed image in the second to-be-processed video. The parameter of the second pixel point can be a pixel value of the second pixel point. The second pixel point is all pixel points in the second to-be-processed image; of course, the second pixel point can also be part of the pixel points in the second to-be-processed image obtained by screening all the pixel points in the second to-be-processed image.

[0075] In the embodiment of the present application, for each second to-be-processed image, a second reference value corresponding to the second to-be-processed image can be determined based on the pixel value of the second pixel point in the second to-be-processed image, and then a second heart rate signal can be generated based on the second reference value corresponding to each second to-be-processed image and the second time corresponding to the collection of each second to-be-processed image. Each second to-be-processed image corresponds to a second reference value.

[0076] It should be noted that when the collection method of the first to-be-processed image and the collection method of the second to-be-processed image are the same, the specific implementation process of generating the first heart rate signal and the specific implementation process of generating the second heart rate signal are the same; when the collection method of the first to-be-processed image and the collection method of the second to-be-processed image are different, the specific implementation process of generating the first heart rate signal and the specific implementation process of generating the second heart rate signal are also different.

[0077] In a feasible implementation manner, the first heart rate signal and the second heart rate signal can be second photoelectric plethysmography (PPG) signals.

[0078] It should be noted that step 102 and step 103 can be executed at the same time; step 102 can also be executed before step 103; step 102 can also be executed after step 103.

[0079] In step 104, based on the first heart rate signal and the second heart rate signal, the blood vessel volume change information and the time difference information of the pulse wave transmission are determined.

[0080] The time difference information represents the difference between the time of blood flow from the heart to the first skin region and the time of blood flow to the second skin region of the target object, and the blood vessel volume change information represents the change of the blood vessel volume when the heart of the target object beats.

[0081] In the embodiments of the present application, the time difference information of the pulse wave transmission can be determined based on the time corresponding to the sample value in the first heart rate signal and the time corresponding to the sample value in the second heart rate signal; the waveforms of the first heart rate signal and the second heart rate signal can be determined, and the blood vessel volume change information can be determined based on the shape of the waveform of the first heart rate signal and the shape of the waveform of the second heart rate signal.

[0082] In step 105, the blood vessel volume change information and the time difference information are processed by the blood pressure detection model to obtain the target blood pressure information of the target object.

[0083] In the embodiments of the present application, the blood vessel volume change information and the time difference information can be input into the blood pressure detection model, and the output result of the blood pressure detection model can be taken as the target blood pressure information of the target object; of course, the output result of the blood pressure detection model can also be adjusted, and the adjusted information can be taken as the target blood pressure information of the target object.

[0084] The information determination method provided by the embodiments of the present application comprises the following steps: a first image acquisition component is used to acquire a first to-be-processed image of a first skin region of a target object, and a second image acquisition component is used to acquire a second to-be-processed image of a second skin region of the target object; the object part corresponding to the first skin region is different from the object part corresponding to the second skin region; a first heart rate signal is generated based on the parameters of a first pixel point in the first to-be-processed image and a first time corresponding to the acquisition of the first to-be-processed image; a second heart rate signal is generated based on the parameters of a second pixel point in the second to-be-processed image and a second time corresponding to the acquisition of the second to-be-processed image; the blood vessel volume change information and the time difference information of pulse wave transmission are determined based on the first heart rate signal and the second heart rate signal; the blood vessel volume change information and the time difference information are processed by a blood pressure detection model to obtain target blood pressure information of the target object; in this way, only the first to-be-processed image and the second to-be-processed image of two different skin regions of the target object, the first time corresponding to the acquisition of the first to-be-processed image, and the second time corresponding to the acquisition of the second to-be-processed image are needed to determine the blood vessel volume change information and the time difference information of pulse wave transmission, and then the blood vessel volume change information and the time difference information of pulse wave transmission are processed by the blood pressure detection model to obtain the target blood pressure information of the target object, without the need for the user to wear a cuff to measure the target blood pressure information in the related art, thereby reducing the complexity of the measurement process of measuring the target blood pressure information and saving the measurement time, and solving the problem of complex process and large time consumption in the related art for measuring blood pressure information.

[0085] Based on the foregoing embodiments, the embodiments of the present application provide an information determination method, applied to an electronic device, as shown in Figure 2 The method comprises the following steps:

[0086] Step 201, a first to-be-processed image of a first skin region of a target object is acquired by a first image acquisition component, and a second to-be-processed image of a second skin region of the target object is acquired by a second image acquisition component.

[0087] The object part corresponding to the first skin region is different from the object part corresponding to the second skin region.

[0088] Step 202, for each frame of the first to-be-processed image, a first pixel value of a pixel point at a first region in the first to-be-processed image is determined.

[0089] The first to-be-processed image is acquired in the case that the first image acquisition component is in contact with the first skin region; the pixel point at the first region in the first to-be-processed image can be understood as all the pixel points in the first region in the first to-be-processed image, or part of the pixel points in the first region.

[0090] In the embodiment of the present application, for each frame of the first to-be-processed image, the image of the first region can be determined from the first to-be-processed image, and the pixel value of the pixel point in the image of the first region is obtained to obtain the first pixel value. Wherein, the electronic device can obtain the target size parameter, and segment the first to-be-processed image based on the target size parameter to obtain the image of the first region. Wherein, the target size parameter can be pre-set.

[0091] In a feasible implementation manner, as shown in Figure 3 The first to-be-processed image can be collected by placing the finger of the target object on the rear camera of the mobile phone; and the image of the first region can be a rectangular region image segmented from the center region of the first to-be-processed image.

[0092] Step 203, obtaining the value of the target color channel from the first pixel value, and performing operation on the value of the target color channel to obtain the first reference value of the first to-be-processed image.

[0093] Wherein, the first pixel value can include the value of the red channel, the value of the green channel and the value of the blue channel.

[0094] In the embodiment of the present application, the target color channel can be pre-set; for a plurality of first pixel values of each frame of the first to-be-processed image, the value of the target color channel can be obtained from each first pixel value, and the plurality of values of the target color channel are operated to obtain the first reference value of the frame of the first to-be-processed image.

[0095] In a feasible implementation manner, for i pixel points in the first region of the Nth frame of the first to-be-processed image, the value of the red channel of the i pixel points can be obtained, and the average of the i values of the red channel is obtained to obtain the first reference value of the Nth frame of the first to-be-processed image; of course, the sum of the i values of the red channel can also be obtained to obtain the first reference value of the Nth frame of the first to-be-processed image. Wherein, i is a positive integer.

[0096] Step 204, generating the first heart rate signal based on the first reference value and the first time.

[0097] Wherein, the first reference value and the first time correspond to each other.

[0098] In the embodiment of the present application, the first corresponding relationship between the first reference value corresponding to the first to-be-processed image and the first time corresponding to the first to-be-processed image, and the order of the plurality of first times can be used to generate the first heart rate signal.

[0099] In a feasible implementation, the number of the first to-be-processed images is 500, the 500 first times can be sorted to obtain sorted first times, and the first heart rate signal is determined based on the sorted first times and the first correspondence; the first heart rate signal is a discrete signal with time as an independent variable; the sample value in the first heart rate signal is a first reference value. The first heart rate signal has 500 points, and each point represents a first reference value corresponding to a different first time.

[0100] In step 205, the face region of the target object in each second to-be-processed image is regionally divided to obtain a plurality of different second regions.

[0101] The second skin region includes the face region; the second to-be-processed image is collected in a case where the second image collection component is not in contact with the second skin region.

[0102] In the embodiments of the present application, for each second to-be-processed image, the second to-be-processed image can be subjected to face detection to determine a face region from the second to-be-processed image, and the face region can be subjected to face key point detection to obtain parameters of a plurality of key points of the target face, and the face region can be divided into a plurality of different second regions according to the parameters of the plurality of key points. The parameters of the key points can be coordinates of the key points.

[0103] In a feasible implementation, as shown in Figure 3 The first to-be-processed image can be collected when the face region of the person is within the range of view of the front camera of the mobile phone without being in contact with the front camera of the mobile phone; for each second to-be-processed image, the face region of the target object in the second to-be-processed image is determined, and the face region is detected using a 68-point key point model of the face, so that 68 key point coordinates of the face region can be finally obtained, and then the face region can be divided into a forehead region, a left cheek region, a right cheek region, a nose region, and a chin region based on the 68 key point coordinates; the plurality of different second regions include the forehead region, the left cheek region, the right cheek region, the nose region, and the chin region.

[0104] In step 206, a reference signal corresponding to each type of second region is determined based on the second pixel value and the second time of each type of second region.

[0105] In the embodiments of the present application, the plurality of different second regions can be understood as a plurality of different types of second regions; each second to-be-processed image corresponds to a plurality of types of second regions; the second pixel value is a pixel value of all pixel points in each type of second region, or a pixel value of part of the pixel points.

[0106] In a possible implementation, the number of the categories of the second regions is 5; wherein the first category of the second regions can be the forehead region, the second category of the second regions can be the left cheek region, the third category of the second regions can be the right cheek region, the fourth category of the second regions can be the nose region, and the fifth category of the second regions can be the chin region.

[0107] In the embodiment of the present application, for the plurality of second to-be-processed images, the second pixel value of each second region in the jth category of second regions and the second time corresponding to the second to-be-processed image where each second region is located are obtained to determine the reference signal corresponding to the jth category of second regions; wherein the reference signal can be an RGB color mode (RGB) signal; and j is a positive integer.

[0108] In a possible implementation, the number of the second to-be-processed images is 500, and for the forehead region, the reference signal corresponding to the forehead region can be generated based on the pixel value (i.e., the second pixel value) of each pixel point in the forehead region of the 500 second to-be-processed images and the second time corresponding to each of the 500 second to-be-processed images; wherein the reference signal corresponding to the forehead region can be represented by RGB1; similarly, the reference signal corresponding to the left cheek region, the reference signal corresponding to the right cheek region, the reference signal corresponding to the nose region, and the reference signal corresponding to the chin region can be obtained in the above manner. The reference signal corresponding to the left cheek region, the reference signal corresponding to the right cheek region, the reference signal corresponding to the nose region, and the reference signal corresponding to the chin region are represented by RGB2, RGB3, RGB4, and RGB5 respectively.

[0109] It should be noted that step 206 can be implemented by steps A1-A2.

[0110] Step A1, determining the second reference value of each second region based on the plurality of second pixel values of each second region.

[0111] In the embodiment of the present application, for each second region in each second to-be-processed image, the plurality of second pixel values of the plurality of pixel points in the second region can be operated to obtain the second reference value of the second region.

[0112] In a possible implementation, for each second region, the plurality of second pixel values in the second region are averaged (i.e., the values of the red channel, the green channel, and the blue channel in the plurality of pixel values are averaged) to obtain the second reference value of the second region; wherein the second reference value includes the value corresponding to the red channel (i.e., the average value of the values of the red channel), the value corresponding to the green channel (i.e., the average value of the values of the green channel), and the value corresponding to the blue channel (i.e., the average value of the values of the blue channel).

[0113] Step A2, obtaining a reference signal corresponding to each type of second region based on the plurality of second reference values of each type of second region and the second time.

[0114] In the embodiments of the present application, the reference signal corresponding to each type of second region is constructed based on the second correspondence relationship between the second reference value of each second region in the Jth type of second region and the second time corresponding to the second region in the second to-be-processed image, and the sequence of the plurality of second times.

[0115] In a feasible implementation manner, the number of the second to-be-processed images is 500, the 500 second times can be sorted to obtain the sorted second times, and the second reference signal is determined based on the sorted second times and the second correspondence relationship; wherein the second reference signal is a discrete signal with time as the independent variable; the sample value in the second reference signal is the second reference value. Wherein, the second reference signal has 500 points, and each point represents the second reference value corresponding to different second time.

[0116] Step 207, processing each reference signal to obtain a second heart rate signal.

[0117] In the embodiments of the present application, for each reference signal, the reference signal can be denoised, and the denoised signal can be converted to obtain a second heart rate signal; in this way, the noise of the second heart rate signal can be reduced, so as to improve the accuracy of determining the target blood pressure information of the target object based on the second heart rate signal.

[0118] In a feasible implementation manner, the first heart rate signal and the second heart rate signal can both be PPG signals.

[0119] It should be noted that step 207 can be implemented by steps B1-B3:

[0120] Step B1, filtering processing each reference signal, and synthesizing each sample value in each processed reference signal in a color channel to obtain a target value.

[0121] In the embodiments of the present application, for each reference signal, the reference signal can be filtered to filter out the noise in the reference signal to obtain a processed reference signal; wherein each sample value in each processed reference signal includes a value corresponding to a red channel, a value corresponding to a green channel and a value corresponding to a blue channel, the value corresponding to the red channel, the value corresponding to the blue channel and the value corresponding to the blue channel can be operated to obtain a target value.

[0122] It should be noted that B1 can be implemented by steps b1-b5:

[0123] b1, determining a third reference value corresponding to the first color channel, a fourth reference value corresponding to the second color channel, and a fifth reference value corresponding to the third color channel.

[0124] In the embodiments of the present application, the third reference value, the fourth reference value, and the fifth reference value can be pre-set, and can also be calculated based on the sample values in the reference signal. The first color channel can be a red channel; the second color channel can be a green channel; and the third color channel can be a blue channel.

[0125] In a feasible implementation manner, the value corresponding to the red channel in each sample value of the plurality of sample values of the reference signal can be obtained, and the plurality of values corresponding to the red channel are averaged to obtain the third reference value. The value corresponding to the green channel in each sample value of the plurality of sample values of the reference signal can be obtained, and the plurality of values corresponding to the green channel are averaged to obtain the fourth reference value. The value corresponding to the blue channel in each sample value of the reference signal can be obtained, and the plurality of values corresponding to the blue channel are averaged to obtain the fifth reference value.

[0126] b2, for each sample value, determining a first value based on the value corresponding to the first color channel in the sample value, the third reference value, and a weight coefficient of the first color channel.

[0127] The first color channel is a red channel.

[0128] In the embodiments of the present application, the weight coefficient of the first color channel can be pre-set, or can be obtained by analyzing the sample values in the reference signal. The value corresponding to the first color channel, the third reference value, and the weight coefficient of the first color channel can be operated to obtain the first value.

[0129] In a feasible implementation manner, the first reference value can be obtained by dividing the value corresponding to the first color channel by the third reference value, and the first value can be obtained by multiplying the first reference value by the weight coefficient of the first color channel.

[0130] b3, for each sample value, determining a second value based on the value corresponding to the second color channel in the sample value and the fourth reference value.

[0131] The second color channel is a green channel.

[0132] In the embodiments of the present application, the value corresponding to the second color channel and the fourth reference value can be operated to obtain the second value.

[0133] In a feasible implementation manner, the second value can be obtained by dividing the value corresponding to the second color channel by the fourth reference value.

[0134] b4. For each sample value, determining a third value based on a value corresponding to the third color channel in the sample value, the fifth reference value, and a weight coefficient of the third color channel.

[0135] The third color channel is a blue color channel.

[0136] In the embodiments of the present application, the weight coefficient of the third color channel can be pre-set or obtained by analyzing the sample values in the reference signal; the value corresponding to the third color channel, the fifth reference value, and the weight coefficient of the third color channel can be operated to obtain the third value.

[0137] In a feasible implementation manner, the second reference value can be obtained by dividing the value corresponding to the third color channel by the fifth reference value, and the third value can be obtained by multiplying the second reference value and the weight coefficient of the third color channel.

[0138] b5. Operating the first value, the second value, and the third value to determine a target value corresponding to the sample value.

[0139] In the embodiments of the present application, for each sample value in the reference signal, the first value, the second value, and the third value can be obtained, then the first value and the third value can be summed to obtain a first sum value, and then the second value can be subtracted from the first sum value to obtain the target value corresponding to the sample value; each sample value corresponds to a target value.

[0140] Step B2. Determining a candidate signal based on the plurality of target values and the second time.

[0141] The number of candidate signals is the same as the number of reference signals; the candidate signal is obtained by processing the reference signal, and the reference signal and the candidate signal correspond one by one.

[0142] In the embodiments of the present application, the candidate signal can be generated based on a third correspondence relationship between the target value corresponding to the second to-be-processed image and the second time corresponding to the second to-be-processed image, and the order of the plurality of second times.

[0143] In a feasible implementation manner, the plurality of second times are sorted, and the candidate signal is determined based on the sorted times and the third correspondence relationship; the candidate signal is a discrete signal with time as the independent variable, and the sample value corresponding to the independent variable is the target value.

[0144] Step B3. Decomposing the candidate signal, and determining a signal of a target frequency band from the decomposed signal to obtain a second heart rate signal.

[0145] The target frequency band can be pre-set.

[0146] In the embodiments of the present application, for each candidate signal, the candidate signal can be subjected to modal decomposition to obtain signals of multiple frequency bands, and the signal of a target frequency band is determined from the signals of the multiple frequency bands, and the signal of the target frequency band is taken as the second heart rate signal corresponding to the candidate signal, so as to improve the accuracy of the determined second heart rate signal, and further improve the accuracy of the target blood pressure information of the target object determined.

[0147] In a feasible implementation manner, the candidate signal can be subjected to adaptive decomposition by using an empirical mode decomposition technique to obtain signals of multiple different frequency bands.

[0148] Step 208: respectively adjusting the first heart rate signal and the second heart rate signal to obtain a first target signal and a second target signal.

[0149] The number of the second heart rate signals is multiple.

[0150] In the embodiments of the present application, the peak values and the valley values in the first heart rate signal are filtered to obtain the first target signal, and the peak values and the valley values in the second heart rate signal are filtered to obtain the second target signal, so as to remove the erroneous peak values and the valley values in the first heart rate signal and remove the erroneous peak values and the valley values in the second heart rate signal, and the accuracy of the blood vessel volume change information and the time difference information determined based on the first target signal and the second target signal is improved.

[0151] It should be noted that step 208 can be implemented by steps C1-C3:

[0152] Step C1: performing interpolation processing on the first heart rate signal based on the first time and the target time interval to obtain a first interpolation signal.

[0153] In the embodiments of the present application, the target time corresponding to each sample value in the first heart rate signal can be determined based on the target time interval and the first time corresponding to each sample value, and the first time corresponding to each sample value is replaced based on the target time to obtain the first interpolation signal, so that the interval between the times corresponding to two adjacent sample values in the first interpolation signal is the same, and the accuracy of the target blood pressure information of the target object determined subsequently is further improved.

[0154] In a feasible implementation manner, if the target time interval is 10 ms, the first time corresponding to the first sample value in the first heart rate signal is t1, and the first time corresponding to the second sample value is t2, then the first time t1 corresponding to the first sample value is taken as the reference, the target time corresponding to the second sample value is t1+10 ms, and t1+10 ms is used to replace t2.

[0155] Step C2: performing interpolation processing on the second heart rate signal based on the second time and the target time interval to obtain a second interpolation signal.

[0156] It should be noted that the interpolation processing is performed on the first heart rate signal and the second heart rate signal respectively, which can strengthen the first heart rate signal and the second heart rate signal, so that the target blood pressure information determined based on the first heart rate signal and the second heart rate signal is more stable.

[0157] In the embodiment of the present application, the processing process of the interpolation processing on the second heart rate signal based on the second time and the target time interval can refer to the description of step C1, which will not be repeated here in the embodiment of the present application.

[0158] Step C3, based on the fifth time corresponding to the peak value and the sixth time corresponding to the valley value in the first interpolation signal, and the seventh time corresponding to the peak value and the eighth time corresponding to the valley value in the second interpolation signal, respectively adjusting the first interpolation signal and the second interpolation signal to obtain the first target signal and the second target signal.

[0159] In the embodiment of the present application, the first interpolation signal can be initially adjusted based on the fifth time corresponding to the peak value and the sixth time corresponding to the valley value in the first interpolation signal, and then the first interpolation signal can be adjusted again based on the fifth time corresponding to the peak value and the sixth time corresponding to the valley value in the first interpolation signal, and the seventh time corresponding to the peak value and the eighth time corresponding to the valley value in the second interpolation signal, to obtain the first target signal; the second interpolation signal can be initially adjusted based on the seventh time corresponding to the peak value and the eighth time corresponding to the valley value in the second interpolation signal, and then the second interpolation signal can be adjusted again based on the fifth time corresponding to the peak value and the sixth time corresponding to the valley value in the first interpolation signal, and the seventh time corresponding to the peak value and the eighth time corresponding to the valley value in the second interpolation signal, to obtain the second target signal.

[0160] It should be noted that the initial adjustment and the secondary adjustment both refer to the adjustment of the peak value and the valley value. Among them, the initial adjustment of the first interpolation signal and the second interpolation signal refers to deleting the erroneous peak value and valley value existing in the first interpolation signal and the second interpolation signal; the secondary adjustment of the first interpolation signal refers to deleting the non-corresponding peak value and the non-corresponding valley value in the first interpolation signal and the second interpolation signal.

[0161] In a feasible implementation, when the first interpolation signal is initially adjusted, for the n th peak value in the first interpolation signal, whether the n th peak value is deleted can be determined according to a second sub-time difference between a fifth time corresponding to the (n-1) th peak value and a fifth time corresponding to the n th peak value, and a third sub-time difference between the fifth time corresponding to the n th peak value and a fifth time corresponding to the (n+1) th peak value; when the second sub-time difference is greater than a second target time difference threshold, and the third sub-time difference is greater than the second target time difference threshold, it is determined that the n th peak value is an error peak value, and the n th peak value needs to be deleted; for the n th valley value in the first interpolation signal, whether the n th valley value is deleted can be determined according to a fourth sub-time difference between a sixth time corresponding to the (n-1) th valley value and a sixth time corresponding to the n th peak value, and a fifth sub-time difference between the sixth time corresponding to the n th valley value and a sixth time corresponding to the (n+1) th valley value; when the fourth sub-time difference is greater than a third target time difference threshold, and the fifth sub-time difference is greater than the third target time difference threshold, it is determined that the n th valley value is an error peak value, and the n th valley value needs to be deleted; in this way, the error peak value and the error valley value existing in the first interpolation signal are deleted in the above manner to realize the initial adjustment of the first interpolation signal; the implementation process of the initial adjustment of the second interpolation signal is basically similar to that of the initial adjustment of the first interpolation signal, and the initial adjustment of the second interpolation signal will not be described here.

[0162] The first target time difference threshold and the second target time difference threshold are pre-set; the first target time difference threshold and the second target time difference threshold can be the same or different. The first target time difference threshold and the second target time difference threshold can be 50 ms.

[0163] In a feasible implementation, the number of the second heart rate signals is 5, which are PPGh1, PPGh2, PPGh3, PPGh4 and PPGh5, and the number of the first heart rate signal is 1, which is represented by PPGf. The first heart rate signal and each second heart rate signal can be compared to delete the non-corresponding peak value and the non-corresponding valley value in the first heart rate signal and each second heart rate signal; for example, the n th sample value in the PPGf signal is a peak value, but the n th sample value in the PPGh1 is a valley value, then the n th sample value in the PPGf signal needs to be deleted, and the n th sample value in the PPGh1 also needs to be deleted; when the n th sample value in the PPGf signal is a peak value, the n th sample value in the PPGh1 is a peak value, or the n th sample value in the PPGf signal is a valley value, and the n th sample value in the PPGh1 is a valley value, then the n th sample value in the PPGf signal is retained, and the n th sample value in the PPGh1 is also retained.

[0164] Step 209, determining time difference information based on the third time corresponding to the first target sample value in the first target signal and the fourth time corresponding to the second target sample value in the second target signal.

[0165] In the embodiment of the present application, the time corresponding to the same type of sample value in the first target sample value and the second target sample value can be operated to obtain an operation result, and the time difference information can be determined based on the multiple operation results. The types of sample values are two types; the first type is a peak value, and the second type is a valley value.

[0166] In a feasible implementation manner, the first target sample value includes a peak value and a valley value in the first target signal; the second target sample value includes a peak value and a valley value in the second target signal; the types of sample values are two types; the first type is a peak value, and the second type is a valley value.

[0167] It should be noted that step 209 can be implemented by steps D1-D3.

[0168] Step D1, calculating a first time difference between a first sub-time corresponding to a peak value in the first target signal and a third sub-time corresponding to a peak value in the second target signal.

[0169] The peak value in the first target signal has a corresponding relationship with the peak value in the second target signal.

[0170] In the embodiment of the present application, the time difference between the first sub-time and the third sub-time can be determined to obtain the first time difference.

[0171] In a feasible implementation manner, the number of the first target signal is 1, and the number of the second target signal is 5; for each second target signal, the time difference between the first sub-time corresponding to the i th peak value in the first target signal and the third sub-time corresponding to the i th peak value in the second target signal can be calculated to obtain the first time difference; in this way, according to this manner of determining the first time difference, multiple first time differences corresponding to multiple pairs of peak values having a corresponding relationship in the first target signal and the second target signal can be obtained.

[0172] Step D2, calculating a second time difference between a second sub-time corresponding to a valley value in the first target signal and a fourth sub-time corresponding to a valley value in the second target signal.

[0173] The valley value in the first target signal has a corresponding relationship with the valley value in the second target signal; the third time includes the first sub-time and the second sub-time; and the fourth time includes the third sub-time and the fourth sub-time.

[0174] In the embodiment of the present application, the time difference between the second sub-time and the fourth sub-time can be determined to obtain the second time difference.

[0175] In a feasible implementation, the number of the first target signals is 1, and the number of the second target signals is 5; for each second target signal, a second time difference between a second sub-time corresponding to an i-th valley in the first target signal and a fourth sub-time corresponding to an i-th valley in the second target signal can be calculated, to obtain a second time difference; in this way, according to the manner of determining the second time difference, a plurality of second time differences corresponding to a plurality of pairs of valleys having a corresponding relationship in the first target signal and the second target signal can be obtained.

[0176] Step D3, performing operation on the plurality of first time differences and the plurality of second time differences to determine time difference information.

[0177] In the embodiments of the present application, the plurality of first time differences and the plurality of second time differences can be summed to obtain a second sum value, and then the second sum value can be averaged to obtain a target mean value, and the target mean value can be taken as the time difference information.

[0178] In a feasible implementation, the number of the first target signals is 1, and the number of the second target signals is 5; for each second target signal, a second time difference between a second sub-time corresponding to an i-th valley in the first target signal and a fourth sub-time corresponding to an i-th valley in the second target signal can be calculated, to obtain a second time difference; in this way, according to the manner of determining the second time difference, a plurality of second time differences corresponding to a plurality of pairs of valleys having a corresponding relationship in the first target signal and the second target signal can be obtained.

[0179] It should be noted that according to the Moens-Korteweg (M-K) formula, the propagation time of the pulse wave in a blood vessel satisfies the following formula (1):

[0180]

[0181] Where t is the pulse wave propagation time, L is the blood vessel length, K is the blood vessel parameter (constant), p is the blood density, D is the blood vessel diameter, E is the Young's modulus of elasticity, and h is the blood vessel wall thickness. Where D and h are relatively fixed, and the relationship between the Young's modulus of elasticity E and the blood pressure information P satisfies the following formula (2):

[0182] E = E0 * e αp Formula (2)

[0183] Where E0 is the Young's modulus of elasticity when the pressure is 0, p is the blood vessel parameter; and a is a constant.

[0184] In the case of the blood vessel length L, the relationship between the pulse wave transmission time and the blood pressure information can be established, as shown in the following formula (3):

[0185]

[0186] Wherein, PTT (Pulse trail time) represents the transmission time of the pulse wave; p represents the blood pressure information; a is a constant. Obviously, it can be seen that the transmission time of the pulse wave is positively correlated with the blood pressure information, so the time difference information of the transmission of the pulse wave of the target object can be calculated as a basis for determining the target blood pressure information of the target object, so as to improve the accuracy of the determined target blood pressure information.

[0187] Step 210, determine the first waveform corresponding to the first target signal and the second waveform corresponding to the second target signal, and determine the blood vessel volume change information based on the first waveform and the second waveform.

[0188] In the embodiments of the present application, the first shape feature can be determined based on the shape of the first waveform, and the second shape feature can be determined based on the shape of the second waveform, and the blood vessel volume change information can be determined based on the first shape feature and the second shape feature.

[0189] In a possible implementation, for each first waveform, the first height difference between the peak value and the valley value in the first waveform, the third time difference between the time corresponding to the peak value and the time corresponding to the first derivative peak value, the second height difference between the peak value and the first derivative peak value, and the peak area can be calculated; wherein the peak area refers to the area of the concave between two peaks in the first waveform; the blood vessel volume change information includes the first height difference, the second height difference, the third time difference and the peak area.

[0190] It should be noted that the heart will cause blood vessel volume change when pumping blood, resulting in changes in the amount of light absorbed by the skin, generating PPG signals, and different blood pressure information will cause different amplitudes of blood vessel volume change when pumping blood, thereby causing changes in the shape of the PPG signal.

[0191] Wherein, the PPG signal can be divided into the superposition of direct current component and alternating current component, the direct current component is mainly caused by the light absorption of the part of the skin tissue except blood, and the alternating current part is mainly caused by the blood change caused by the contraction and diastole of the heart, according to the Beer-Lambert law, the light transmission intensity I s in the systole period and the light transmission intensity I d satisfy the formula:

[0192]

[0193]

[0194] Wherein, I0 is the incident light intensity, α DC = ε * c is the product of the absorption coefficient of the other tissues outside the blood (i.e. the direct current component of PPG) and the material concentration of the other tissues outside the blood (such as subcutaneous fat), dDC is the optical path of the PPG DC component, a b is the absorption coefficient of blood multiplied by the concentration, D s is the absorption coefficient of blood multiplied by the concentration, D d is the absorption coefficient of blood multiplied by the concentration, D

[0195]

[0196] wherein, Ad is the difference between the systolic and diastolic diameters of the blood vessel; it can be seen that the AC characteristics of the PPG signal can reflect the change in the diameter of the blood vessel when the heart beats, and further reflect the blood pressure information. The greater the blood pressure, the greater the contraction; the greater the blood pressure, the greater the difference; the greater the pressure, the greater the blood vessel; therefore, the AC characteristics of the PPG signal, i.e. the blood vessel volume change information, can be considered as a parameter for determining the target blood pressure information of the target object.

[0197] It should be noted that step 210 can be implemented by steps E1-E3:

[0198] Step E1, segmenting the first target signal based on adjacent peak values in the first target signal to obtain a plurality of first sub-signals, and segmenting the second target signal based on adjacent peak values in the second target signal to obtain a plurality of second sub-signals.

[0199] In the embodiments of the present application, the first target signal can be analyzed, and based on the adjacent two peak values in the first target signal, the first target signal is segmented to obtain the first sub-signal corresponding to the adjacent two peak values in the first target signal; the second target signal can be analyzed, and based on the adjacent peak values in the second target signal, the second target signal is segmented to obtain the second sub-signal corresponding to the adjacent two peak values in the second target signal.

[0200] In a feasible implementation manner, taking the first target signal as an example, if the number of peak values in the first target signal is 10, then a first sub-signal is determined by the adjacent two peak values starting from the first peak value, and then 5 first sub-signals can be obtained.

[0201] Step E2, determining the waveform of each first sub-signal to obtain a first waveform.

[0202] In the embodiments of the present application, the first waveform corresponding to each first sub-signal can be generated according to each first sub-signal; or the waveform corresponding to each first sub-signal can be obtained from the waveform of the first target signal based on each first sub-signal, and the first waveform is obtained.

[0203] Step E3, determining the waveform of each second sub-signal to obtain a second waveform.

[0204] In the embodiments of the present application, the second waveform corresponding to each segment of the second sub-signal can be generated according to each segment of the second sub-signal; or the waveform corresponding to the second target signal is generated first, and then the waveform corresponding to each segment of the second sub-signal is obtained from the waveform of the second target signal based on each segment of the second sub-signal to obtain the second waveform.

[0205] In step 211, attribute information of the target object is obtained.

[0206] In the embodiments of the present application, the attribute information of the target object can be input by the target object to the electronic device; or the electronic device can send an acquisition instruction for acquiring the attribute information of the target object to another device, and the attribute information of the target object is sent by the other device to the electronic device based on the acquisition instruction; wherein the attribute information of the target object is stored in the other device. The attribute information of the target object can represent the physical characteristics of the target object.

[0207] In a feasible implementation manner, the target object is a person, and the attribute information of the target object can be at least one of the height of the target object, the weight of the target object, and the age of the target object.

[0208] In step 212, the blood vessel volume change information, the time difference information, and the attribute information are processed by the blood pressure detection model to obtain target initial blood pressure information of the target object.

[0209] In the embodiments of the present application, the blood vessel volume change information, the time difference information, and the attribute information can be input into the blood pressure detection model, so that the blood pressure detection model can analyze the blood vessel volume change information, the time difference information, and the attribute information to output the target initial blood pressure information of the target object; in this way, the physical characteristics of the target object are considered, and the blood vessel volume change information and the time difference information are combined to determine the target initial blood pressure information, thereby improving the accuracy of the determined target initial blood pressure information. The blood vessel volume change information can be represented by FPPGs; the time difference information can be represented by FPTT; and the attribute information can be represented by Ffix.

[0210] It should be noted that the working environment information, the working time length, and the time length of using the electronic device of the target object can also be obtained, and at least one of the working environment information, the working time length, and the time length of using the electronic device is combined with the blood vessel volume change information, the time difference information, and the attribute information to be input into the blood pressure detection model to obtain the target initial blood pressure information of the target object, so as to determine the target initial blood pressure information from multiple dimensions to improve the accuracy of the determined target initial blood pressure information.

[0211] In a feasible implementation manner, the electronic device can be a mobile phone used by the target object.

[0212] In the embodiment of the present application, sample blood vessel volume change information, sample time difference information and sample attribute information of the sample object can be acquired, and a target classifier is used to perform model training based on the sample blood vessel volume change information, the sample time difference information and the sample attribute information, so as to obtain a blood pressure detection model.

[0213] In a feasible implementation, the target classifier includes a support vector machine (SVM) classifier and a random forest classifier.

[0214] In step 213, an association between the blood pressure information of the target object and the initial blood pressure information is obtained, and the target blood pressure information of the target object is determined based on the target initial blood pressure information and the association.

[0215] In the embodiment of the present application, the calibration blood pressure information of the target object and the initial reference blood pressure information of the target object can be acquired, and based on the calibration blood pressure information and the initial reference blood pressure information, the association between the blood pressure information of the target object and the initial blood pressure information is determined. The calibration blood pressure information can be the real blood pressure information of the target object at the nearest time before the first use of the blood pressure detection model. When the target object uses the blood pressure detection model for the first time, the initial attribute information, the determined initial blood vessel volume change information and the initial time difference information can be input into the blood pressure detection model, and the information output by the blood pressure detection model can be used as the initial reference blood pressure information.

[0216] The association between the blood pressure information of the target object and the initial blood pressure information can be determined according to the difference information between the initial reference blood pressure information and the calibration blood pressure information. The association represents the relationship between the real blood pressure information of the target object and the blood pressure information output by the blood pressure detection model.

[0217] In the embodiment of the present application, the real blood pressure information of the target object corresponding to the target initial blood pressure information can be determined based on the target initial blood pressure information and the association, so as to obtain the target blood pressure information of the target object. Further, the accuracy of the determined target blood pressure information is improved.

[0218] It should be noted that the information determination method in the embodiment of the present application can be deployed on a mobile phone, a notebook computer and other smart devices equipped with two cameras, without the need to use a special blood pressure measuring device. The software and hardware combination is used to capture the finger and the face by using the two cameras respectively, and the quality of the first to-be-processed image for the finger and the second to-be-processed image for the face is improved by adjusting the shooting parameters such as exposure time, white balance and light and other parameters, so as to further improve the signal-to-noise ratio of the determined first heart rate signal and the second heart rate signal.

[0219] Furthermore, the simultaneous capture of finger and face images perfectly aligns with users' mobile phone usage habits, improving the comfort of blood pressure measurement and enhancing user engagement. By employing dual cameras to acquire PPG signals from both the face and finger, and comparing and analyzing the PPG signals from the face and finger, the target blood pressure information for the subject is obtained, improving the accuracy of the determined target blood pressure information. The incorporation of vascular volume change information based on PTT characteristics enhances the robustness of blood pressure measurement.

[0220] It should be noted that the same steps and contents as in other embodiments can be described in other embodiments, and will not be repeated here.

[0221] The information determination method provided in the embodiments of this application only needs to determine the vascular volume change information and pulse wave transmission time difference of the target object based on the first image to be processed and the second image to be processed of two different skin areas of the target object, the first time corresponding to the acquisition of the first image to be processed, and the second time corresponding to the acquisition of the second image to be processed. Then, the vascular volume change information and pulse wave transmission time difference are processed by the blood pressure detection model to obtain the target blood pressure information of the target object. This eliminates the need for the user to measure the target blood pressure information by wearing a cuff, which reduces the complexity of the measurement process and saves measurement time. It solves the problem that the process of measuring blood pressure information is complicated and time-consuming in the related technologies.

[0222] Based on the foregoing embodiments, embodiments of this application provide an electronic device that can be applied to... Figures 1-2 In the information determination method provided in the corresponding embodiment, refer to Figure 4 As shown, the electronic device may include a first image acquisition component 31, a second image acquisition component 32, and a processor 33;

[0223] The first image acquisition unit 31 is used to acquire a first image to be processed for a first skin region of the target object;

[0224] The second image acquisition unit 32 is used to acquire a second image to be processed for a second skin region of the target object; wherein the object part corresponding to the first skin region is different from the object part corresponding to the second skin region.

[0225] Processor 33 is used to generate a first heart rate signal based on the parameters of the first pixel in the first image to be processed and the first time corresponding to the acquisition of the first image to be processed;

[0226] The processor 33 is also used to generate a second heart rate signal based on the parameters of the second pixel in the second image to be processed and the second time corresponding to the acquisition of the second image to be processed;

[0227] The processor 33 is further configured to determine the blood vessel volume variation information and the time difference information of pulse wave transmission based on the first heart rate signal and the second heart rate signal.

[0228] The processor 33 is further configured to process the blood vessel volume variation information and the time difference information by using a blood pressure detection model to obtain target blood pressure information of the target object.

[0229] In other embodiments of the present application, the processor 33 is specifically configured to perform the following steps:

[0230] The first target signal and the second target signal are obtained by adjusting the first heart rate signal and the second heart rate signal respectively.

[0231] The time difference information is determined based on a third time corresponding to a first target sample value in the first target signal and a fourth time corresponding to a second target sample value in the second target signal.

[0232] The first waveform corresponding to the first target signal and the second waveform corresponding to the second target signal are determined, and the blood vessel volume variation information is determined based on the first waveform and the second waveform.

[0233] In other embodiments of the present application, the processor 33 is specifically configured to perform the following steps:

[0234] The first interpolation signal is obtained by performing interpolation processing on the first heart rate signal based on the first time and the target time interval.

[0235] The second interpolation signal is obtained by performing interpolation processing on the second heart rate signal based on the second time and the target time interval.

[0236] The first target signal and the second target signal are obtained by adjusting the first interpolation signal and the second interpolation signal respectively based on a fifth time corresponding to a peak value in the first interpolation signal and a sixth time corresponding to a valley value, and a seventh time corresponding to a peak value in the second interpolation signal and an eighth time corresponding to a valley value.

[0237] In other embodiments of the present application, the processor 33 is specifically configured to perform the following steps:

[0238] The first time difference between a first sub-time corresponding to a peak value in the first target signal and a third sub-time corresponding to a peak value in the second target signal is calculated.

[0239] a second time difference between a second sub-time corresponding to a valley in the first target signal and a fourth sub-time corresponding to a valley in the second target signal, wherein the peak in the first target signal and the peak in the second target signal have a corresponding relationship, the valley in the first target signal and the valley in the second target signal have a corresponding relationship, the third time includes the first sub-time and the second sub-time, and the fourth time includes the third sub-time and the fourth sub-time;

[0240] performing operation on the plurality of first time differences and the plurality of second time differences to determine the time difference information.

[0241] In other embodiments of the present application, the processor 33 is specifically configured to perform the following steps:

[0242] segmenting the first target signal based on adjacent peaks in the first target signal to obtain a plurality of first sub-signals, and segmenting the second target signal based on adjacent peaks in the second target signal to obtain a plurality of second sub-signals;

[0243] determining a waveform of each first sub-signal to obtain a first waveform;

[0244] determining a waveform of each second sub-signal to obtain a second waveform.

[0245] In other embodiments of the present application, the processor 33 is specifically configured to perform the following steps:

[0246] for each frame of the first to-be-processed image, determining a first pixel value of a pixel point at the first region from the first to-be-processed image, wherein the first to-be-processed image is collected under the condition that the first image acquisition component is in contact with the first skin region;

[0247] obtaining a value of a target color channel from the first pixel value, and performing operation on the value of the target color channel to obtain a first reference value of the first to-be-processed image;

[0248] generating a first heart rate signal based on the first reference value and the first time.

[0249] In other embodiments of the present application, the processor 33 is specifically configured to perform the following steps:

[0250] performing region division on a face region of the target object in each frame of the second to-be-processed image to obtain a plurality of different second regions; wherein the second skin region includes the face region; wherein the second to-be-processed image is collected under the condition that the second image acquisition component is not in contact with the second skin region;

[0251] determining a reference signal corresponding to each type of second region based on a second pixel value of each type of second region and a second time;

[0252] process each reference signal to obtain a second heart rate signal;

[0253] In other embodiments of the present application, the processor 33 is specifically configured to perform the following steps:

[0254] determine a second reference value of each second region based on a plurality of second pixel values of each second region;

[0255] obtain a reference signal corresponding to each type of second region based on a plurality of second reference values of each type of second region and the second time.

[0256] In other embodiments of the present application, the processor 33 is specifically configured to perform the following steps:

[0257] filter each reference signal, and synthesize each value in each filtered reference signal in a color channel to obtain a target value;

[0258] determine a candidate signal based on a plurality of target values and the second time;

[0259] decompose the candidate signal, and determine a signal of a target frequency band from the decomposed signal to obtain a second heart rate signal.

[0260] In other embodiments of the present application, the processor 33 is specifically configured to perform the following steps:

[0261] obtain attribute information of the target object;

[0262] process the blood vessel volume change information, the time difference information, and the attribute information through a blood pressure detection model to obtain target initial blood pressure information of the target object;

[0263] obtain a correlation relationship between the blood pressure information of the target object and the initial blood pressure information, and determine target blood pressure information of the target object based on the target initial blood pressure information and the correlation relationship.

[0264] The electronic device provided in the embodiments of the present application only needs to determine the blood vessel volume change information and the time difference of the pulse wave transmission based on the first to-be-processed image and the second to-be-processed image of two different skin regions of the target object, the first time corresponding to the acquisition of the first to-be-processed image, and the second time corresponding to the acquisition of the second to-be-processed image, and then processes the blood vessel volume change information and the time difference of the pulse wave transmission through a blood pressure detection model to obtain the target blood pressure information of the target object. The user does not need to wear a cuff to measure the target blood pressure information in the related art, which reduces the complexity of the measurement process of measuring the target blood pressure information and saves the measurement time, and solves the problem of complex process and large time consumption in measuring the blood pressure information in the related art.

[0265] Based on the foregoing embodiments, the embodiments of the present application provide a computer readable storage medium storing one or more programs, which can be executed by one or more processors to implement Figures 1-2 The steps of the information determination method provided by the corresponding embodiments.

[0266] It should be noted that the computer readable storage medium described above can be a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Programmable Read-Only Memory (EPROM), an Electrically Erasable Programmable Read-Only Memory (EEPROM), a Ferromagnetic Random Access Memory (FRAM), a Flash Memory, a magnetic surface memory, an optical disc, a Compact Disc Read-Only Memory (CD-ROM), or the like. It can also be various electronic devices including one or any combination of the above memories, such as a mobile phone, a computer, a tablet device, a personal digital assistant, and the like.

[0267] It should be noted that in this document, the term "comprising" or "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or apparatus including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to such a process, method, article or apparatus. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of other identical elements in the process, method, article or apparatus including the element.

[0268] The serial numbers of the embodiments of the present application described above are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0269] Through the above description of the embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment method can be realized by means of software and a necessary general hardware platform, and of course, it can also be realized by hardware, but in many cases, the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium (such as a ROM / RAM, a magnetic disk, or an optical disk) and includes a plurality of instructions for causing a terminal device (which can be a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the methods described in the various embodiments of the present application.

[0270] The present application is described with reference to the flowcharts and / or block diagrams according to the methods, devices (systems), and computer program products of the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the computer or other programmable data processing devices generate a device that realizes the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one or more flows and / or blocks Figure 1 The functions specified in one or more flows and / or blocks

[0271] These computer program instructions can also be stored in a computer-readable memory that can guide the computer or other programmable data processing devices to work in a specific way, so that the instructions stored in the computer-readable memory produce a manufactured product that includes instruction devices that realize the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one or more flows and / or blocks Figure 1 The functions specified in one or more flows and / or blocks

[0272] These computer program instructions can also be loaded into a computer or other programmable data processing device, so that a series of operation steps are performed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide a device for realizing the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one or more flows and / or blocks Figure 1 The functions specified in one or more flows and / or blocks

[0273] The above is only the preferred embodiments of the present application, and does not limit the patent scope of the present application, and any equivalent structure or equivalent flow transformation made by using the contents of the specification and drawings, or directly or indirectly applied to other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. An information determination method, wherein, The method comprises: acquiring, by a first image acquisition component, a first to-be-processed image of a first skin region of a target object, and acquiring, by a second image acquisition component, a second to-be-processed image of a second skin region of the target object; wherein the first skin region corresponds to a different part of the target object from the second skin region; generating a first heart rate signal based on parameters of first pixel points in the first to-be-processed image and a first time corresponding to acquisition of the first to-be-processed image; generating a second heart rate signal based on parameters of second pixel points in the second to-be-processed image and a second time corresponding to acquisition of the second to-be-processed image; determining blood vessel volume variation information and time difference information of pulse wave transmission based on the first heart rate signal and the second heart rate signal; processing the blood vessel volume variation information and the time difference information by a blood pressure detection model to obtain target blood pressure information of the target object; wherein the determination of the blood vessel volume variation information based on the first heart rate signal and the second heart rate signal comprises: adjusting the first heart rate signal and the second heart rate signal respectively to obtain a first target signal and a second target signal; segmenting the first target signal based on adjacent peak values in the first target signal to obtain a plurality of first sub-signals, and segmenting the second target signal based on adjacent peak values in the second target signal to obtain a plurality of second sub-signals; determining a waveform of each first sub-signal to obtain a first waveform; and determining a waveform of each second sub-signal to obtain a second waveform; for each first waveform and each second waveform, calculating a first height difference between a peak value and a valley value, a third time difference between a time corresponding to the peak value and a time corresponding to a first derivative peak value, a second height difference between the peak value and the first derivative peak value, and a peak area; and the blood vessel volume variation information comprises the first height difference, the second height difference, the third time difference, and the peak area.

2. The method of claim 1, wherein, the determination of the blood vessel volume variation information and the time difference information of pulse wave transmission based on the first heart rate signal and the second heart rate signal comprises: determining the time difference information based on a third time corresponding to a first target sample value in the first target signal and a fourth time corresponding to a second target sample value in the second target signal.

3. The method of claim 2, wherein, the adjustment of the first heart rate signal and the second heart rate signal respectively to obtain the first target signal and the second target signal comprises: performing interpolation processing on the first heart rate signal based on the first time and a target time interval to obtain a first interpolation signal; performing interpolation processing on the second heart rate signal based on the second time and the target time interval to obtain a second interpolation signal; adjusting the first interpolation signal and the second interpolation signal respectively based on a fifth time corresponding to a peak value in the first interpolation signal and a sixth time corresponding to a valley value, and a seventh time corresponding to a peak value in the second interpolation signal and an eighth time corresponding to a valley value, to obtain the first target signal and the second target signal.

4. The method of claim 2, wherein, The time difference information is determined based on a third time corresponding to a first target sample value in the first target signal and a fourth time corresponding to a second target sample value in the second target signal. A first time difference between a first sub-time corresponding to a peak value in the first target signal and a third sub-time corresponding to a peak value in the second target signal is calculated. A second time difference between a second sub-time corresponding to a valley value in the first target signal and a fourth sub-time corresponding to a valley value in the second target signal is calculated, wherein the peak value in the first target signal has a corresponding relationship with the peak value in the second target signal, and the valley value in the first target signal has a corresponding relationship with the valley value in the second target signal, the third time includes the first sub-time and the second sub-time, and the fourth time includes the third sub-time and the fourth sub-time. The time difference information is determined by performing operations on a plurality of first time differences and a plurality of second time differences.

5. The method of claim 1, wherein, The first heart rate signal is generated based on a parameter of a first pixel point in the first to-be-processed image and a first time corresponding to acquisition of the first to-be-processed image. For each frame of the first to-be-processed image, a first pixel value of a pixel point at a first region in the first to-be-processed image is determined, wherein the first to-be-processed image is acquired in a case where the first image acquisition component is in contact with the first skin region. A value of a target color channel is obtained from the first pixel value, and the value of the target color channel is operated to obtain a first reference value of the first to-be-processed image. The first heart rate signal is generated based on the first reference value and the first time.

6. The method of claim 1, wherein, The second heart rate signal is generated based on a parameter of a second pixel point in the second to-be-processed image and a second time corresponding to acquisition of the second to-be-processed image. A plurality of different second regions are obtained by performing region division on a face region of the target object in each frame of the second to-be-processed image, wherein the second skin region includes the face region, and the second to-be-processed image is acquired in a case where the second image acquisition component is not in contact with the second skin region. A reference signal corresponding to each type of second region is determined based on a second pixel value of each type of second region and the second time. Each reference signal is processed to obtain the second heart rate signal. The reference signal corresponding to each type of second region is determined based on the second pixel value of each type of second region and the second time, including: A second reference value of each second region is determined based on a plurality of second pixel values of each second region. The reference signal corresponding to each type of second region is obtained based on a plurality of second reference values of each type of second region and the second time.

7. The method of claim 6, wherein, Each reference signal is filtered, and each sample value in each processed reference signal is synthesized in a color channel to obtain a target value. A candidate signal is determined based on a plurality of target values and the second time. ​ The candidate signal is decomposed, and a signal of a target frequency band is determined from the decomposed signal to obtain the second heart rate signal.

8. The method of claim 1, wherein, The blood pressure detection model is used to process the blood vessel volume change information and the time difference information to obtain target blood pressure information of the target object. Attribute information of the target object is obtained. The blood pressure detection model is used to process the blood vessel volume change information, the time difference information, and the attribute information to obtain target initial blood pressure information of the target object. A correlation relationship between blood pressure information and initial blood pressure information of the target object is obtained, and target blood pressure information of the target object is determined based on the target initial blood pressure information and the correlation relationship.

9. An electronic device, comprising: The electronic device comprises a first image acquisition component, a second image acquisition component, and a processor. The first image acquisition component is configured to acquire a first to-be-processed image of a first skin region of a target object. The second image acquisition component is configured to acquire a second to-be-processed image of a second skin region of the target object. The processor is configured to generate a first heart rate signal based on parameters of a first pixel point in the first to-be-processed image and a first time when the first to-be-processed image is acquired. The processor is further configured to generate a second heart rate signal based on parameters of a second pixel point in the second to-be-processed image and a second time when the second to-be-processed image is acquired. The processor is further configured to determine blood vessel volume change information and time difference information of pulse wave transmission based on the first heart rate signal and the second heart rate signal. The processor is further configured to process the blood vessel volume change information and the time difference information by using a blood pressure detection model to obtain target blood pressure information of the target object. The processor is specifically configured to adjust the first heart rate signal and the second heart rate signal respectively to obtain a first target signal and a second target signal, segment the first target signal based on adjacent peak values in the first target signal to obtain a plurality of first sub-signals, and segment the second target signal based on adjacent peak values in the second target signal to obtain a plurality of second sub-signals, determine a waveform of each first sub-signal to obtain a first waveform, determine a waveform of each second sub-signal to obtain a second waveform, calculate, for each first waveform and each second waveform, a first height difference between a peak value and a valley value in each waveform, a third time difference between a time corresponding to the peak value and a time corresponding to a first derivative peak value, a second height difference between the peak value and the first derivative peak value, and a peak area, and the blood vessel volume change information comprises the first height difference, the second height difference, the third time difference, and the peak area.

Images