TWM564431U - Non-contact heartbeat rate measurement apparatus - Google Patents

Abstract

A non-contact heartbeat rate measurement apparatus includes an image sensor and a computing module. The image sensor is configured to capture a plurality of facial images continuously. The computing module is coupled to the image sensor. The computing module is configured to select target areas from the facial images, obtain a heartbeat signal according to color variances of pixels in the target areas, and perform spectrum analysis on the heartbeat signal to obtain a heartbeat spectrum. The heartbeat spectrum includes heartbeat signal strength values under different frequencies. The computing module is configured to calculate a signal-noise ratio of the heartbeat spectrum, calculate a face-moving frequency according to the facial images, and select a frequency from the heartbeat spectrum as an output heartbeat frequency according to the signal-noise ratio and the face-moving frequency.

Description

Non-contact heart rate measuring device

The present disclosure relates to a measurement system and measurement method, and more particularly to a non-contact heart rate measurement system and a non-contact heart rate measurement method.

A lot of important health information can be obtained by measuring heartbeat. In general, the conventional heartbeat measurement method uses a contact type heartbeat measurement method, that is, a sensor patch is directly attached to the subject to measure the heartbeat signal of the subject. However, the conventional contact type heartbeat measurement method always makes the subject feel inconvenient and uncomfortable.

A non-contact heartbeat measurement system of the present disclosure includes an image sensor, a target area selection module, a heartbeat signal calculation module, a spectrum analysis module, a shake detection module, and a heartbeat peak selection module. The image sensor is used to continuously capture a plurality of facial images. The target area selection module is used to select a target area from each of the face images. The heartbeat signal calculation module is configured to calculate a color difference amount of each pixel in the target area of the successively captured facial images to obtain a heartbeat signal. The spectrum analysis module is used for spectrum analysis of the heartbeat signal, and To the heartbeat spectrum, the heartbeat spectrum includes a plurality of heartbeat signal strength values at a plurality of frequencies, and calculates a signal-to-noise ratio and entropy of the heartbeat spectrum. The shaking detection module is configured to detect a facial shaking frequency according to the facial images. When the heartbeat peak selection module determines that the signal noise ratio is compared and the entropy is higher than the threshold value and the first peak frequency having the highest signal intensity value in the heartbeat spectrum is similar to the face shaking frequency, the heartbeat peak selection module is composed of the heartbeat spectrum. A second peak frequency having a local highest signal strength value is selected as the output heartbeat frequency in the local frequency band.

A non-contact heartbeat measurement method of the present disclosure includes selecting a target region from each of a plurality of facial images. The heartbeat signal is obtained according to the color difference amount of each pixel in the target area in the face image captured successively. Perform spectrum analysis on the heartbeat signal and obtain a heartbeat spectrum. The heartbeat spectrum includes a plurality of heartbeat signal strength values at a plurality of frequencies, and calculates a signal-to-noise ratio and entropy of the heartbeat spectrum. The face shaking frequency is calculated from the face image. When the noise ratio and the entropy are higher than the threshold value and the first peak frequency having the highest signal intensity value of the whole domain in the heartbeat frequency is similar to the face shaking frequency, the second local frequency band having the highest signal strength value is selected from the local frequency band of the heartbeat frequency. The peak frequency is used as the output heartbeat frequency.

The present disclosure further provides a non-contact heartbeat measuring device that includes an image sensor and an arithmetic module. The image sensor is used to continuously capture a plurality of facial images. The computing module is coupled to the image sensor, and the computing module is configured to select a target area from each of the facial images, and according to the successively captured facial images, the pixels in the target area The color difference quantity obtains a heartbeat signal, performs spectrum analysis on the heartbeat signal, and obtains a heartbeat spectrum, and the heartbeat spectrum includes a plurality of heartbeat signal strengths at a plurality of frequencies. And calculating a signal noise ratio of the heartbeat spectrum, calculating a facial shaking frequency according to the facial images, and selecting the plurality of frequencies of the heartbeat spectrum according to the signal noise ratio and the facial shaking frequency One of the frequencies acts as an output heartbeat frequency.

100‧‧‧ Non-contact heartbeat measurement system

110,310‧‧‧Image sensor

121‧‧‧Target area selection module

121a‧‧‧Feature point coordinate detection unit

121b‧‧‧Target area selection unit

122‧‧‧ heartbeat signal calculation module

123‧‧‧Spectrum Analysis Module

124‧‧‧Shake detection module

125‧‧‧ Heartbeat Peak Selection Module

126‧‧‧Heartbeat Change Protection Module

127‧‧‧Adaptive filter

130, 330‧‧‧ Output interface

300‧‧‧ Non-contact heart rate measuring device

320‧‧‧ Computing Module

340‧‧‧Power supply module

341‧‧‧ battery module

342‧‧‧Power switch button

343‧‧‧Charging hole

350‧‧‧Fixed modules

AI‧‧‧ overall image

EFP‧‧‧ Eye feature point coordinates

FHR‧‧‧Filtered heartbeat signal

FI‧‧‧Face image

FVF‧‧‧Face shaking frequency

GFB‧‧‧ global frequency band

HR‧‧‧ heartbeat signal

HS‧‧‧ heartbeat spectrum

MFP‧‧‧ mouth feature point coordinates

OHR‧‧‧Output heartbeat frequency

OHR1‧‧‧ first output heart rate

OHR2‧‧‧second output heartbeat frequency

OHR3‧‧‧ third output heart rate

PF1‧‧‧ first peak frequency

PF2‧‧‧second peak frequency

PF3‧‧‧ third peak frequency

PFB‧‧‧local band

SNR‧‧‧ signal noise ratio

TR‧‧‧Target area

M100‧‧‧ Non-contact heartbeat measurement method

S101~S110‧‧‧Steps

The above and other objects, features, advantages and embodiments of the present disclosure will become more apparent and understood. The description of the drawings is as follows: FIG. 1 is a non-contact heartbeat according to an embodiment of the present disclosure. A functional block diagram of the measurement system.

2 is a flow chart of a non-contact heartbeat measurement method according to an embodiment of the present disclosure.

3 is a further functional block diagram of a target area selection module of a contactless heartbeat measurement system in accordance with an embodiment of the present disclosure.

FIG. 4 is a further flow chart of step S120 of the non-contact heartbeat measurement method according to an embodiment of the present disclosure.

Figure 5A is a schematic illustration of a facial image shown in accordance with one embodiment of the present disclosure.

Fig. 5B is a schematic diagram of detecting the feature point coordinates of the mouth and the coordinates of the eye feature points in the facial image shown in Fig. 5A.

FIG. 5C is a schematic diagram of the target area selected from the face image of the mouth feature point coordinates and the eye feature point coordinates shown in FIG. 5B.

Figure 6A is a schematic illustration of a heartbeat signal shown in accordance with one embodiment of the present disclosure.

FIG. 6B is a schematic diagram of the filtered heartbeat signal after the heartbeat signal shown in FIG. 6A is filtered by a band pass filter.

Figure 6C is a schematic diagram of the heartbeat spectrum obtained by analyzing the filtered heartbeat signal spectrum shown in Fig. 6B.

FIG. 7A is a diagram showing the first peak frequency as the output heartbeat frequency, according to an embodiment of the present disclosure.

FIG. 7B is a schematic diagram showing the second peak frequency as the output heartbeat frequency according to an embodiment of the present disclosure.

Figure 7C is a schematic diagram showing the third peak frequency as the output heartbeat frequency, according to an embodiment of the present disclosure.

Figure 8 is a functional block diagram of a non-contact heartbeat measurement system in accordance with another embodiment of the present disclosure.

FIG. 9A is a front elevational view of a non-contact heartbeat measuring device according to an embodiment of the present disclosure.

Fig. 9B is a side view showing the non-contact type heartbeat measuring device in Fig. 9A.

The embodiments are described in detail below to better understand the present invention, but the embodiments provided are not intended to limit the scope of the present disclosure, and the description of structural operations is not intended to limit the scope thereof. The order of execution, any structure that is recombined by components, produces equal devices, and is covered by this case.

Please refer to Figure 1 and Figure 2. Figure 1 is based on the present disclosure A functional block diagram of the non-contact heartbeat measurement system 100 shown in one embodiment. 2 is a flow chart of a non-contact heartbeat measurement method M100 according to an embodiment of the present disclosure.

In the present embodiment, the non-contact heartbeat measurement system 100 can be used to perform a non-contact heartbeat measurement method M100 for performing non-contact heartbeat measurement, wherein the non-contact heartbeat measurement system 100 includes an image sensor 110, The target area selection module 121, the heartbeat signal calculation module 122, the spectrum analysis module 123, the shake detection module 124, the heartbeat peak selection module 125, and the output interface 130, the non-contact heartbeat measurement method M100 includes steps S101 to Step S110. Image sensor 110 can be an optical sensing element or a camera unit.

In step S101, the image sensor 110 can continuously capture a plurality of facial images FI. Specifically, please refer to FIG. 5A, which is a schematic diagram of a facial image FI according to an embodiment of the present disclosure. The image sensor 110 first captures the user's overall image AI, and uses the face capture technology to capture the facial image FI in the overall image AI, wherein the face extraction technique can be a multi-level convolutional neural network. (convolutional neural network, CNN), but not limited to this.

In an embodiment, the image sensor 110 can be a camera, a video camera, a video recorder, or the like.

In step S102, the target area selection module 121 selects the target area TR from the face image FI.

Further, please refer to FIG. 3, FIG. 4, FIG. 5B and FIG. 5C together. FIG. 3 is a non-connection according to an embodiment of the present disclosure. A further functional block diagram of the target area selection module 121 of the touch heartbeat measurement system 100, and FIG. 4 is a further flow chart of the step S120 of the non-contact heart rate measurement method M100 according to an embodiment of the present disclosure. FIG. 5B is a schematic diagram of detecting the mouth feature point coordinates MFP and the eye feature point coordinates EFP in the face image FI shown in FIG. 5A. FIG. 5C is a schematic diagram of the frame selection target region TR in the face image FI indicated by the mouth feature point coordinate MFP and the eye feature point coordinate EFP shown in FIG. 5B.

As shown in FIG. 3, the target area selection module 121 further includes a feature point coordinate detecting unit 121a and a target area frame selecting unit 121b. As shown in FIG. 4, step S102 further includes step S102a and step S102b.

The feature point coordinate detecting unit 121a detects the mouth feature point coordinate MFP and the eye feature point coordinate EFP (shown in FIG. 5B) on the face image FI in accordance with step S102a. Specifically, the feature point coordinate detecting unit 121a can mark the mouth and the eye in the face image FI according to the mouth and the graphic features of the eye (for example, a specific shape or a specific color), and define the mouth and the coordinates of the eye, respectively. Thereby, the mouth feature point coordinate MFP and the eye feature point coordinate EFP can be defined.

The target area frame selection unit 121b selects the target area TR based on the mouth feature point coordinates MFP and the eye feature point coordinates EFP in accordance with step S102b. Specifically, after the mouth feature point coordinate MFP and the eye feature point coordinate EFP are defined, the target area frame selection unit 121b can select the target area TR in the area between the mouth feature point coordinate MFP and the eye feature point coordinate EFP. For example, the midpoint of the MFP with the mouth feature points at both ends of the mouth is the first midpoint, and the two eye feature points The midpoint of the standard EFP is the second midpoint, and a midpoint between the first midpoint and the second midpoint is selected as a rectangular target area TR, as shown in FIG. 5C. Further, in order to avoid excessive amplitude of the mouth feature point coordinate MFP and the eye feature point coordinate EFP, a low-pass filter is added, so that the target area TR can be correctly displayed.

In step S103, the heartbeat signal calculation module 122 obtains the heartbeat signal HR according to the color difference amount of each pixel in the target area TR of the face image FI captured successively. Specifically, please refer to FIG. 6A, which is a schematic diagram of a heartbeat signal HR according to an embodiment of the present disclosure, wherein the center skip signal HR is expressed in the time domain, that is, the horizontal axis is time. (s), the vertical axis is the intensity (dB).

The heartbeat signal calculation module 122 performs optical flow calculation on the target area TR of the previous moment and the latter moment to obtain the position of each pixel point at the next moment. In detail, the red, green and blue signals of the corresponding pixel points in the target area TR of the previous moment and the last moment are respectively subtracted, and the corresponding differences of red, green and blue can be obtained, respectively, which are red differences. The value dR, the green difference dG, and the blue difference dB; then, the red difference dR, the green difference dG, and the blue difference dB are linearly combined, for example, the red difference dR, the green difference dG And the blue difference dB is multiplied by a specific weight and then added, thereby obtaining two sets of characteristic signals X and Y; finally, the difference of each color of each pixel is averaged to obtain a heartbeat signal HR. In an embodiment, the heartbeat signal HR is used to indicate the amount of change in the difference at different times. In essence, the physical meaning of this signal application is like Photoplethysmography (PPG). The traditional application method requires wearing wearable devices such as heartbeat chest straps and bracelets, and simultaneously enters a specific light source. The intensity of the reflected signal is highly correlated with the blood flow, so the change in the strength of the reflected signal is regarded as a heartbeat signal. The system uses a long-distance non-contact image sensor to extract the heartbeat signal from the small change in the color of the skin to the reflected signal of the ambient light source. In general, when the heartbeat of the human body is contracted and dilated, the microvessels have different blood pressures, and the facial area is a region where the microvessels are densely distributed. The color of the facial region may be minute with the contraction and relaxation of the heartbeat. The non-contact heart rate measurement module 122 of the present invention uses the color change of the face region to detect the heartbeat signal HR.

In step S104, the spectrum analysis module 123 performs spectrum analysis on the heartbeat signal HR, and obtains a heartbeat spectrum HS. The heartbeat spectrum HS includes a plurality of heartbeat signal intensity values at a plurality of frequencies, and calculates a signal noise of the heartbeat spectrum HS. Ratio SNR (signal-to-noise ratio, SNR) and entropy (Entropy). Specifically, please refer to FIG. 6B and FIG. 6C together. FIG. 6B is a schematic diagram of the filtered heartbeat signal FHR after the heartbeat signal HR shown in FIG. 6A is filtered by a band pass filter, and the sixth C is a pair. A schematic diagram of the heartbeat spectrum HS obtained by spectral analysis of the filtered heartbeat signal FHR shown in FIG. 6B.

In detail, the spectrum analysis module 123 filters the heartbeat signal HR as shown in FIG. 6A every half second (about 15 frames) through a band pass filter, and generates a filtered heartbeat signal FHR as shown in FIG. 6B. . Then, the spectrum analysis module 123 performs a fast Fourier transform (FFT) on the filtered heartbeat signal FHR to obtain a heartbeat spectrum HS as shown in FIG. 6C, and the center hop spectrum HS is expressed in the frequency domain. That is, the horizontal axis is the frequency (Hz), the vertical axis is the intensity (dB), and the heartbeat spectrum HS includes a plurality of heartbeat signal intensity values at a plurality of frequencies. Next, the spectrum analysis module 123 calculates the signal of the heartbeat spectrum HS. The noise ratio SNR and entropy, the signal noise ratio SNR is the ratio of the power of signal and the power of noise of the heartbeat spectrum HS. Entropy, a function that describes the state of a system, is often used to calculate the degree of disorder and confusion in a system.

In step S105, the shake detection module 124 detects the face shake frequency FVF based on the face image FI. Specifically, the sway detection module 124 can calculate the sway frequency FVF according to the displacement of the mouth feature point coordinate MFP of the face image FI and the eye feature point coordinate EFP at the previous moment and the subsequent moment, and match the speed. The technique of compensation to reduce the facial sloshing frequency FVF.

In step S106, the heartbeat peak selection module 125 determines whether the signal noise ratio SNR or entropy of the heartbeat spectrum HS is higher than a threshold value, wherein the threshold value can be set according to actual conditions, for example, the SNR threshold can be set to 0.8 and entropy. The threshold can be set to 4.5.

When the signal-to-noise ratio SNR of the heartbeat spectrum HS is higher than the threshold and the entropy is lower than the threshold, for example, the signal-to-noise ratio SNR of the heartbeat spectrum HS is 0.9 and the entropy is 2, indicating that the heartbeat spectrum HS is subjected to the face shaking frequency FVF. The influence is extremely small without affecting the heartbeat spectrum HS, at which point it proceeds to step S107.

In step S107, the heartbeat peak selection module 125 determines whether the first peak frequency PF1 having the highest signal strength of the whole region in the heartbeat spectrum HS is similar to the facial sway frequency FVF.

When the first peak frequency PF1 of the heart rate spectrum HS having the highest signal intensity of the whole region is different from the face sloshing frequency FVF, it indicates that the first peak frequency PF1 having the highest signal strength of the whole region is mainly generated by the heartbeat, instead of the face. The shaking frequency FVF is generated, and at this time, it proceeds to step S108.

In step S108, reference is made to FIG. 7A, which is a schematic diagram showing the first peak frequency PF1 as the output heartbeat frequency OHR according to an embodiment of the present disclosure. That is to say, the heartbeat spectrum HS shown in FIG. 7A is that the signal noise ratio SNR according to the heartbeat spectrum HS is higher than the threshold value and the first peak frequency PF1 having the highest signal strength in the heartbeat spectrum HS is different from the face shaking. The heartbeat spectrum of the condition of the frequency FVF.

Further, it can be observed from FIG. 7A that in the global frequency band GFB, the heartbeat spectrum HS includes a plurality of heartbeat signal strength values at a plurality of frequencies, and the first peak frequency PF1 having the highest frequency signal strength of the whole region can be used as the output heartbeat frequency. OHR is output through output interface 130, such as a projector or display.

In another embodiment, when the first peak frequency PF1 having the highest signal strength of the whole region in the heartbeat spectrum HS is similar to the facial sway frequency FVF, the first peak frequency PF1 having the highest signal strength of the whole region is mainly caused by the shaking of the face. The frequency FVF is generated, and at this time, it proceeds to step S109.

In step S109, reference is made to FIG. 7B, which is a schematic diagram showing the second peak frequency PF2 as the output heartbeat frequency OHR according to an embodiment of the present disclosure. That is to say, the heartbeat spectrum HS shown in FIG. 7B is that the signal noise ratio SNR according to the heartbeat spectrum HS is higher than the threshold value and the first peak frequency PF1 having the highest signal strength in the heartbeat spectrum HS is similar to the face shaking frequency. The heartbeat spectrum of the condition of FVF.

Further, it can be observed from FIG. 7B that in the global frequency band GFB, the heartbeat spectrum HS includes a plurality of heartbeat signal strength values at a plurality of frequencies, and the heartbeat spectrum HS also includes a facial sway frequency FVF having a global maximum signal strength. To avoid directly swaying the face with the highest signal strength in the world, FVF As the output heartbeat frequency OHR, the local frequency band PFB can be set in the global frequency band GFB, wherein the range of the local frequency band PFB is a frequency range conforming to the general heartbeat, that is, 1 (Hz) to 4 (Hz). Thereby, the second peak frequency PF2 having the highest signal strength in the local frequency band PFB can be used as the output heartbeat frequency OHR and output through the output interface 130, such as a projector or a display.

In another embodiment, when the signal-to-noise ratio SNR of the heartbeat spectrum HS is lower than the threshold and the entropy is higher than the threshold, for example, the signal-to-noise ratio SNR of the heartbeat spectrum HS is 0.1, indicating that the heartbeat spectrum HS is subjected to facial shaking. The influence of the frequency FVF is extremely large and affects the heartbeat spectrum HS, at which point it proceeds to step S110.

In step S110, reference is made to FIG. 7C, which is a schematic diagram showing the third peak frequency PF3 as the output heartbeat frequency OHR according to an embodiment of the present disclosure. That is to say, the heartbeat spectrum HS shown in FIG. 7C is a heartbeat spectrum that satisfies the condition that the signal-to-noise ratio SNR of the heartbeat spectrum HS is lower than the threshold value.

Further, it can be observed from FIG. 7C that in the global frequency band GFB, the heartbeat spectrum HS includes a plurality of heartbeat signal strength values at a plurality of frequencies, and the heartbeat spectrum HS also includes a facial sway frequency FVF having a global maximum signal strength. In order to avoid directly erroneously detecting the face sloshing frequency FVF having the highest signal strength of the whole region as the output heartbeat frequency OHR, the local frequency band PFB can be set in the global frequency band GFB, wherein the range of the local frequency band PFB is a frequency range conforming to the general heartbeat, that is, 1 (Hz) ~ 5 (Hz). Thereby, the third peak frequency PF3 having the highest signal strength in the local frequency band PFB will be available as the output heartbeat frequency OHR and output through the output interface 130, such as a projector or a display.

It should be noted that the target area of the non-contact heartbeat measurement system 100 The domain selection module 121, the feature point coordinate detecting unit 121a, the target area frame selection unit 121b, the heartbeat signal calculation module 122, the spectrum analysis module 123, the shake detection module 124, and the heartbeat peak selection module 125 are available in hardware. , software, firmware or a combination thereof.

Referring again to FIG. 8, a functional block diagram of a non-contact heartbeat measurement system 200 in accordance with another embodiment of the present disclosure.

The non-contact type heartbeat measuring system 200 shown in FIG. 8 is substantially the same as the non-contact type heartbeat measuring system 100 shown in FIG. 1, and the difference is the non-contact type heartbeat measuring system shown in FIG. The 200 further includes a heartbeat change protection module 126 and an adaptive filter 127. In order to highlight the differences, the similarities are not repeated.

The heartbeat change protection module 126 is configured to calculate an average value and a standard deviation of the first output heartbeat frequency OHR1 output by the heartbeat peak selection module 125, and add and subtract the average value and the standard deviation to form a boundary value. If the first output heartbeat frequency OHR1 exceeds the boundary value, the boundary value is output to become the second output heartbeat frequency OHR2, and if the first output heartbeat frequency OHR1 does not exceed the boundary value, the first output heartbeat frequency OHR1 is output to become the second output heartbeat frequency. OHR2.

The adaptive filter 127 is configured to eliminate Gaussian noise and measurement error of the second output heartbeat frequency OHR2 outputted from the heartbeat change protection module 126 to output a third output heartbeat frequency OHR3. In the embodiment, the adaptive filter 127 is a Kalman filter, but is not limited thereto.

It should be noted that the target area selection module 121, the feature point coordinate detection unit 121a, the target area frame selection unit 121b, the heartbeat signal calculation module 122, the spectrum analysis module 123, and the target of the contactless heartbeat measurement system 200 The motion detection module 124 and the heartbeat peak selection module 125, the heartbeat change protection module 126, and the adaptive filter 127 can be embodied by hardware, software, firmware, or a combination thereof.

In summary, the heartbeat peak selection module can determine whether the signal noise ratio is higher than the threshold value and whether the first peak frequency having the highest signal intensity value of the whole region is similar to the condition of the face shaking frequency, and correspondingly Selecting the first peak frequency, the second peak frequency or the third peak frequency output heartbeat frequency OHR, thereby achieving the purpose of anti-sloshing heartbeat measurement; furthermore, passing through the heartbeat change protection module and the adaptive filter, thereby The output heartbeat signal is more stable.

Please refer to FIG. 9A and FIG. 9B together. FIG. 9A is a front elevational view of a non-contact heartbeat measuring device 300 according to an embodiment of the present disclosure. FIG. 9B is a side view showing the non-contact type heartbeat measuring device 300 in FIG. 9A.

As shown in FIG. 9A, the non-contact heartbeat measuring device 300 includes an image sensor 310 and an arithmetic module 320. The image sensor 310 can be an optical sensing element or a camera unit. The computing module 320 can be an embedded physiological signal computing module, a processor, a special application integrated circuit, or other equivalent computing circuits. In an embodiment, the computing module 320 implements the target area selection module 121, the heartbeat signal calculation module 122, the spectrum analysis module 123, and the shaking detection described in the previous embodiment through a software, a firmware, or a hardware. The module 124, the heartbeat peak selection module 125, the heartbeat change protection module 126, and the adaptive filter 127 (see FIGS. 1 and 8).

The image sensor 310 is configured to continuously capture a plurality of facial images (please refer to the facial images FI in FIGS. 5A to 5C). Computing module 320 is coupled to image sensor 310. In an embodiment, the operation module 320 can be used to perform the non-contact heartbeat measurement method M100 described in the previous embodiment (refer to FIG. 2 and the related embodiments), and the operation module 320 is from the faces. Each of the image images selects a target region, and obtains a heartbeat signal according to the color difference of each pixel in the target region, and performs spectrum analysis on the heartbeat signal to obtain a heartbeat spectrum. The heartbeat spectrum includes a plurality of heartbeat signal strength values at a plurality of frequencies, and calculates a signal noise ratio of the heartbeat spectrum, and calculates a facial shaking frequency according to the facial images, according to the signal noise ratio and The face shaking frequency selects one of the plurality of frequencies of the heartbeat spectrum as an output heartbeat frequency. In one embodiment, when the signal to noise ratio is higher than the threshold and the first peak frequency having the highest signal strength value in the heartbeat spectrum is similar to the facial shaking frequency, the operation module 320 is configured by the local frequency band of the heartbeat spectrum. The second peak frequency having the local highest signal strength value is selected as the output heartbeat frequency.

The above-mentioned processing flow and technical details of the operation module 320 have been described in detail in steps S101 to S109 (see FIG. 2) of the previous embodiment, and are not described herein.

As shown in FIG. 9A, the non-contact heartbeat measuring device 300 further includes an output interface 330. The output interface 330 is coupled to the computing module 320 and the image sensor 310. The output interface 330 can be used to display the output heartbeat frequency and the These facial images.

As shown in FIG. 9A and FIG. 9B , the non-contact heartbeat measuring device 300 further includes a power supply module 340 , a battery module 341 , and a power switch button 342 . And a charging hole 343. The power supply module 340 and the battery module 341 are disposed in the non-contact heartbeat measuring device 300. In the embodiment of FIGS. 9A and 9B, the power switch button 342 and the charging hole 343 are provided on the side surface of the non-contact type heartbeat measuring device 300. The power switch button 342 is used to switch the switching state of the non-contact heartbeat measuring device 300. The charging hole 343 is used to overlap with a matched power input (for example, a transformer, a power converter, etc., not shown). The power supply module 340 is electrically connected to the battery module 341, the power switch button 342, the image sensor 310, and the computing module 320 for selectively supplying power to the image capturing module and the computing module according to the switch state. .

As shown in FIG. 9A and FIG. 9B , the non-contact heartbeat measuring device 300 further includes a fixing module 350 for fixing the non-contact heartbeat measuring device 300 to an external object (eg, a wall surface, a desktop, a door frame, Columns or other objects, not shown in the figure). The fixing module 350 shown in FIG. 9A and FIG. 9B is a fixing clip having a screw locking structure, but the disclosure is not limited thereto. In practical applications, the fixing module 350 may also be a fixing clip having a spring structure. , snap-fit structure, fixed strap structure, elastic collar or any other mechanical structure that can provide a fixed effect.

Although the present invention has been disclosed in the above embodiments, it is not intended to limit the present invention. Anyone having ordinary knowledge in the technical field can protect the case without any deviation and refinement within the spirit and scope of the present case. The scope is subject to the definition of the scope of the patent application attached.

Claims (5)

A non-contact type heartbeat measuring device includes: an image sensor for continuously capturing a plurality of facial images; and an operation module coupled to the image sensor, the operation module is used for Selecting a target area from each of the facial images, and obtaining a heartbeat signal according to the color difference of each pixel in the target image, and performing spectrum analysis on the heartbeat signal, and Obtaining a heartbeat spectrum, the heartbeat spectrum includes a plurality of heartbeat signal intensity values at a plurality of frequencies, and calculating a signal noise ratio of the heartbeat spectrum, and calculating a facial shaking frequency according to the facial images, according to the signal The noise ratio and the facial shaking frequency are used to select one of the plurality of frequencies of the heartbeat spectrum as an output heartbeat frequency. The contactless heartbeat measuring device of claim 1, wherein the signal to noise ratio is higher than a threshold and the first peak frequency of the global highest signal strength value in the heartbeat spectrum is similar to the face When the frequency is shaken, the computing module selects a second peak frequency having a local highest signal strength value from a local frequency band of the heartbeat spectrum as the output heartbeat frequency. The contactless heartbeat measuring device of claim 1, further comprising: an output interface coupled to the computing module and the image sensor for displaying the output heartbeat frequency and the facial images. The non-contact type heartbeat measuring device of claim 1, further comprising: a battery module; a charging hole; a power switch button for switching a switching state of the non-contact type heartbeat measuring device; and a power source The power module is electrically connected to the battery module, the power switch button, the image sensor, and the computing module for selectively supplying power to the image capturing module and the computing module according to the switch state. The contactless heartbeat measuring device of claim 1, further comprising: a fixing module for fixing the non-contact type heartbeat measuring device to an external object.