KR20220143307A - Heart rate measurement method and device using color image - Google Patents

Abstract

The present invention relates to a technique for measuring the heart rate based on a color image. According to one embodiment, provided is a device for measuring the heart rate using a color image, which may include: a collection unit for collecting color image signals; a region of interest extraction unit for extracting a region of interest from the collected color image signals; and a heart rate estimation unit for estimating the heart rate by using a color change for each of color information consisting of red, green, and blue colors of the extracted color image of the region of interest.

Description

Heart rate measurement method and device using color image

The present invention relates to a technique for measuring a heartbeat based on a color image, and more particularly, to a technical idea of estimating a heartbeat using only a color change in a region of interest obtained by imaging an image.

Heart rate, or heart rate, is the number of heartbeats per unit time, and is a number generally expressed in beats per minute (bpm). Heart rate can vary depending on life activities, such as physical exercise or sleeping, as the body absorbs oxygen and releases carbon dioxide.

Heart rate measurements are used by health care professionals to diagnose and check medical conditions. It also monitors heart rate for athletes to get maximum efficiency from their training. In general, heart rate is the inverse relationship of the time interval (RR interval) between the R points and the R points in the P-Q-R-S type waveform of the pulse waveform measured using an electrocardiography (ECG) device (only 1 minute is the standard unit) .)

Your heart rate is measured at the heartbeat of your body. This pulse rate can be measured by pressing the surface with the index and middle fingers at any location in the body where there are arteries. Your thumb is not used to measure someone else's heart rate. This is because a strong force may interfere with accurately measuring the target pulse.

Heart rate can be measured at various points in the body.

For example, the radial artery in the ventral area below the thumb, the ulnar artery, the carotid artery, the medial or brachial artery of the elbow, the femoral artery, and the medial malleolus of the foot. It can be measured in the posterior tibial artery and in the middle of the dorsal area (dorsalis pedis).

On the other hand, a more sophisticated method of determining the pulse uses an electrocardiogram (ECG, EKG). Continuous ECG monitoring is used in many medical applications. Commercial heart rate monitoring also exists. The signal is sent from the receiver on the wrist to the monitor. Heart rate monitoring provides a continuous and accurate measurement, and can be used during exercise where manual measurement is difficult or inconvenient, such as when using the hands. Another way to measure your heart rate is to record your body's vibrations.

Despite the existence of various contact-type measurement technologies that measure heart rate, it is a very important factor in monitoring the health of the body. the current situation.

Korean Patent No. 2215557 "Camera-based heart rate measurement method and system using face color and tremor" Korea Patent No. 1787828 "Heart rate measurement system using skin color filter"

An object of the present invention is to provide a technology for measuring heart rate in a non-contact manner using continuous image information of color images.

The present invention uses a minute color change between individual color information (red, green, blue) constituting a color image in order to extract a heartbeat, so it is converted into an image of normalized brightness using the brightness information of the image to achieve brightness compensation and Bayer It aims to have more accurate non-contact measurement performance by minimizing the error due to brightness by estimating the heartbeat using only the actual red, green, and blue measurement colors of the pattern.

An object of the present invention is to measure heartbeats from adults to infants even with general color video information rather than high resolution.

To an apparatus for measuring a heartbeat using a color image according to an embodiment, a collection unit for collecting a color image signal, a region of interest extractor for extracting a region of interest from the collected color image signal, and a color image of the extracted region of interest With respect to , a process of estimating a heartbeat using color changes for each of the color information consisting of red, green, and blue, and converting the extracted luminance signal into a normalized luminance signal, a preprocessing process using filtering, and the color It may include a heartbeat estimator for applying at least one process among processes for applying Linear Discrimination Analysis (LDA) in relation to information.

The heartbeat measuring apparatus using a color image according to an embodiment calculates the brightness signal for the extracted region of interest (ROI), measures individual red, green, and blue values, and converts it into a signal of normalized brightness, , a compensator for compensating for green and blue values may be further included.

The heartbeat measuring apparatus using a color image according to an embodiment further includes a preprocessing unit that preprocesses at least one of the collected color image or the extracted region of interest using a 2D high pass filter. may include

The heart rate estimator according to an embodiment may estimate the heart rate by applying Linear Discrimination Analysis (LDA) that considers a difference between the average values of the color changes for each of the color information.

The heartbeat estimator according to an embodiment serializes each of the color information as a function of time and then applies the LDA (Linear Discrimination Analysis) to each window of a predetermined time, and each color information to which the LDA is applied is determined according to time. The heart rate cycle may be estimated by separating and extracting signals having similar or different values.

According to one embodiment, it is possible to provide a technology for measuring the heart rate in a non-contact manner using continuous image information of a color image.

According to an embodiment, in order to extract a heartbeat, a minute color change between individual color information (red, green, blue) constituting a color image is used. By estimating the beat, it is possible to minimize the error caused by the brightness.

According to an exemplary embodiment, heartbeats from adults to infants may be measured with sufficient resolution even with general color video information rather than high resolution.

1 is a view for explaining a heartbeat measuring apparatus using a color image according to an embodiment.
FIG. 2 is a view for explaining an entire one-frame image taken at a low resolution and an image selected as an ROI.
3 is a diagram for explaining high-pass filtered red, green, and blue images.
4A is a view for explaining a change in image brightness when measuring 500 frames.
4B is a diagram for explaining a time-domain conversion signal of red, green, and blue color information.
4C is a diagram for explaining a heart rate cycle extraction signal using the LDA.
FIG. 4D is a diagram illustrating the result of the autocorrelation function of FIG. 4C.
4E is a diagram for explaining brightness-corrected red, green, and blue signals.
4F is a diagram for explaining a heart rate cycle extraction signal using the LDA.
FIG. 4G is a diagram showing the result of the autocorrelation function of FIG. 4F.
5 is a view for explaining a heartbeat measurement method using a color image according to an embodiment.
6 is a diagram for explaining an N-frame heartbeat extraction image sequence window signal.

Specific structural or functional descriptions of the embodiments according to the concept of the present invention disclosed herein are only exemplified for the purpose of explaining the embodiments according to the concept of the present invention, and the embodiment according to the concept of the present invention These may be embodied in various forms and are not limited to the embodiments described herein.

Since the embodiments according to the concept of the present invention may have various changes and may have various forms, the embodiments will be illustrated in the drawings and described in detail herein. However, this is not intended to limit the embodiments according to the concept of the present invention to specific disclosed forms, and includes changes, equivalents, or substitutes included in the spirit and scope of the present invention.

Terms such as first or second may be used to describe various elements, but the elements should not be limited by the terms. The above terms are used only for the purpose of distinguishing one component from other components, for example, without departing from the scope of rights according to the concept of the present invention, a first component may be named as a second component, Similarly, the second component may also be referred to as the first component.

When a component is referred to as being “connected” or “connected” to another component, it may be directly connected or connected to the other component, but it is understood that other components may exist in between. it should be On the other hand, when it is said that a certain element is "directly connected" or "directly connected" to another element, it should be understood that the other element does not exist in the middle. Expressions describing the relationship between elements, for example, “between” and “between” or “directly adjacent to”, etc. should be interpreted similarly.

The terminology used herein is used only to describe specific embodiments, and is not intended to limit the present invention. The singular expression includes the plural expression unless the context clearly dictates otherwise. In the present specification, terms such as "comprise" or "have" are intended to designate that the described feature, number, step, operation, component, part, or combination thereof exists, and includes one or more other features or numbers, It should be understood that the possibility of the presence or addition of steps, operations, components, parts or combinations thereof is not precluded in advance.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms such as those defined in a commonly used dictionary should be interpreted as having a meaning consistent with the meaning in the context of the related art, and should not be interpreted in an ideal or excessively formal meaning unless explicitly defined in the present specification. does not

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. However, the scope of the patent application is not limited or limited by these embodiments. Like reference numerals in each figure indicate like elements.

1 is a view for explaining a heartbeat measuring apparatus 100 using a color image according to an embodiment.

The heart rate measuring apparatus 100 using a color image according to an embodiment may measure the heart rate in a non-contact manner using continuous information of a general color image. In particular, in order to extract a heartbeat, a minute color change between individual color information (red, green, blue) constituting a color image is used. It can minimize errors due to brightness and has sufficient resolution, so it can measure heartbeats from adults to infants even with general color video information.

To this end, the heartbeat measuring apparatus 100 using a color image according to an embodiment may include a collecting unit 110 , an ROI extracting unit 120 , and a heartbeat estimating unit 130 .

The collection unit 110 according to an embodiment may collect a color image signal.

Also, the extractor 120 according to an embodiment may extract a region of interest from the collected color image signals.

The resolution of the camera may be related to the area of the photographed area selected for heart rate measurement and the number of pixels of the camera allocated to the area (ROI: Region of Interest). For example, even with a low-resolution camera of 640 X 480 pixels, when measuring an ROI of 5 X 5 cm ² on a full screen of 20 X 20 cm ² , 138 X 138 pixels are allocated, so it can have a sufficient resolution of 0.13 mm ² /pixel, so it can have a normal color With video information, it is possible to measure heart rate from adults to infants.

Since the heartbeat estimator 130 according to an embodiment uses a minute color change between individual color information (red, green, blue) constituting a color image to extract a heartbeat, normalized using brightness information of the image By estimating the heartbeat by converting it into an image of brightness, errors due to brightness can be minimized.

The heartbeat estimator 130 according to an embodiment may use the following [Equation 1] for normalized brightness.

[Equation 1]

Normalized BRIGHTNESS = 0.21 * RED + 0.72 * GREEN + 0.07 * BLUE

The heartbeat estimator 130 according to an embodiment extracts a region of interest (ROI) from an image signal of a camera that takes a color image, calculates a brightness signal, measures the values of individual red, green, and blue signals, and normalizes them. converted into a signal of the specified brightness.

Appropriate brightness based on experience is defined as a normalized value of 0.65 - 0.75 and can be used to compensate for red, green, and blue values.

Specifically, the heartbeat estimator 130 according to an embodiment may estimate the heartbeat by using color changes for each of color information composed of red, green, and blue with respect to the extracted color image of the region of interest. have.

A typical adult's heart rate is 60-100 bpm (beat per minute), a child over 1 year old has a heart rate of 70-110 bpm, and a child's heart rate is 100-150 bpm.

Therefore, if the maximum 180bpm is 3Hz, it is possible to measure the heart rate from adults to infants even with a general camera of 15fps (frame per second = 15Hz).

However, since the frame rate of the camera is related to the precision of the heart rate measurement, it can have sufficient measurement accuracy of 5bpm for 15fps, 2bpm for 30fps, and 1bpm when using a 60fps camera.

In particular, the heart rate estimator 130 according to an embodiment may estimate the heart rate by applying Linear Discrimination Analysis (LDA) in consideration of a difference between a covariance between color changes for each color information and an average value.

In addition, the heartbeat estimator 130 according to an embodiment serializes each color information as a function of time and then applies LDA (Linear Discrimination Analysis) to each window of a predetermined time, and each color information to which the LDA is applied is displayed in time. Accordingly, the heart rate cycle may be estimated by separating and extracting signals having similar or different values. Also, the heart rate estimator 130 may estimate the heart rate by using the heart rate cycle estimated in this way.

In general, each color of red, green, and blue has a different degree of absorption and reflection by the skin depending on the wavelength. In particular, red is absorbed by hemoglobin in blood vessels, so the amount of change is small when estimating blood flow. The heartbeat estimator 130 according to an exemplary embodiment makes it possible to better use these different response characteristics, that is, characteristics having different measurement values for each color. That is, the heartbeat estimator 130 according to an embodiment considers the difference between the average values without using a general PCA (Principal Component Analysis) or ICA (Independent Component Analysis), and the direction of the Projection Vector is more With the LDA (Linear Discrimination Analysis) method that finds the difference between different signals, it is possible to estimate the different signals, that is, the heart rate, between the red, green, and blue information.

The heartbeat estimator 130 according to an embodiment serializes each color information as a function of time and applies LDA (Linear Discrimination Analysis) to each window of a predetermined time, and each color information to which the LDA is applied is displayed in time. Accordingly, the heart rate cycle can be estimated by separating and extracting signals having similar or different values within the reference.

After serializing the filtered red, green and blue color information as a time function, LDA (Linear Discrimination Analysis) is used every window (eg, 5 seconds) of a certain time to determine whether each color information has similar or different values over time. The heart rate cycle can be estimated by separating and extracting the signals. A signal representing heart rate information can be obtained directly from the peak-to-peak distance on the time axis, or the extracted signal can be easily found from the first peak value using an autocorrelation function.

The LDA method has some advantages over PCA or ICA using a simple Eigen vector direction because the solution of the following conditions is obtained by considering the red, green, and blue signal itself and the difference in the average value between signals.

[Equation 2]

argmax(W) = W ^T S _B W / W ^T SW W

In [Equation 2], S _B may represent a dispersion value between red, green, and blue, and SW may represent a dispersion value of red, green, and audio signals.

In addition, the heartbeat measuring apparatus 100 using a color image according to an embodiment calculates the brightness signal for the extracted region of interest (ROI), measures individual red, green, and blue values, and converts it into a normalized brightness signal. Accordingly, the compensator 140 may further include individual red, green, and blue values.

In relation to the region of interest (ROI), the diameter of the arteries distributed in the exposed body parts, such as the face and back of the hand, is about 1.5 to 2.5 mm in adults and half the size of adults in infants and children. Since about 100,000 - 160,000 km of blood vessels are distributed inside, the heart rate is extracted using only color information (Red (Red), Green (Green), Blue (Blue)) that changes according to the blood flow of the arteries distributed under the exposed skin. can do.

In addition, the heartbeat measuring apparatus 100 using a color image according to an embodiment uses a 2D high pass filter with respect to at least one of the collected color image or the extracted region of interest. It may further include a pre-processing unit 150 for pre-processing.

For example, since the preprocessor 150 basically uses color information that changes according to the blood flow, successive frame images can be pre-processed using a two-dimensional high-pass filter so that the information about the amount of change individually can be well represented. have.

According to an embodiment, the two-dimensional high-pass filter may be implemented as shown in Equation 3 below.

[Equation 3]

In [Equation 3],

+C _ij is a positive coefficient, -C _ij is a negative coefficient, and +δ is a value of the corresponding position in the Bayer Pattern form of the color image, not an interpolation value, but a highly correlated measurement value measured directly by the pixel, so it is a general 0 value Instead, we can use a high-pass filter that considers these values as small weights. That is, the weight of the smallest value, +δ, can be applied to the value of the nearest pixel (the position of the nearest measurement pixel).

The 2D high-pass filter may be used by extending the order or size of the filter according to the size of the ROI or the resolution of the image.

Meanwhile, the central processing unit 160 may be further included for overall control of each component or for transmitting signals.

FIG. 2 is a view for explaining an entire one-frame image taken at a low resolution and an image selected as an ROI.

In particular, the first frame 210 of FIG. 2 represents an entire one-frame image captured at a low resolution of 960 x 540 pixels and 30 fps using a smartphone.

In particular, FIG. 3 is an 890 x 50 pixel image selected as an ROI, and a shaded portion may be included to examine the effect of brightness correction and noise.

Next, FIG. 3 is a diagram 300 for explaining high-pass filtered red, green, and blue images.

Among the images indicated by reference numeral 300, reference numeral 310 denotes a red image, reference numeral 320 denotes a green image, and reference numeral 330 denotes a blue image.

Red, green, and blue colors are absorbed and reflected differently by the skin depending on the wavelength. Especially, red is absorbed by hemoglobin in blood vessels, so the amount of change is small when estimating blood flow. As different response characteristics for these red, green, and blue images, the difference between average values can be measured without using general PCA (Principal Component Analysis) or ICA (Independent Component Analysis) to better utilize characteristics with different color measurement values. In consideration of this, a Linear Discrimination Analysis (LDA) method may be used.

The LDA (Linear Discrimination Analysis) method finds the difference between signals whose projection vector direction is more different, and can estimate the heart rate of different signals between red, green, and blue information. .

4A is a diagram 410 for explaining a change in image brightness when measuring 500 frames.

Specifically, FIG. 4A shows the change in brightness of 500 images, and the method of calculating the brightness is as follows.

For example, if a color image consists of R, G, and B three color information and each color has a resolution of 8 bits, each color may have a Numeric value in the range of 0 to 255. Therefore, if this is normalized, the value of each color is 0 to 1, so brightness can be calculated as in the following [Equation 4].

[Equation 4]

max = 0.21*1(R_max) + 0.72*1(G_max) + 0.07*1(B_max)=1

4B is a diagram 420 for explaining a time domain conversion signal of red, green, and blue color information. Also, FIG. 4C is a diagram 430 illustrating a heartbeat cycle extraction signal using LDA, and FIG. 4D is a diagram 440 illustrating the result of the autocorrelation function of FIG. 4C.

Meanwhile, FIG. 4E is a diagram 450 for explaining brightness-corrected red, green, and blue signals.

4e is a brightness-corrected RGB signal, where the Y-axis is a value obtained by multiplying the normalized measurement signal by the correction value (R_norm*BR_comp), and the X-axis is the frame number of the image (time axis: 1/Frame, 2/Frame…or frame number). indicates.

The method of correcting the above value for the change in brightness of each red, green, and blue color can be interpreted as reducing each maximum value of [Equation 4] to 65 to 75%.

In other words, the purpose of the brightness correction is not to limit the maximum brightness, but to brighten the dark case and slightly darken it if it is too bright so that the brightness of the image is always 　0.65 - 0.75　.

4F is a diagram 460 for explaining a heart rate cycle extraction signal using the LDA.

The Y-axis of FIG. 4f is the second dominant signal component from the three input signals 　 red, green, and blue, 　 an independent signal component representing the change according to the blood flow, and the magnitude is the 　W matrix multiplication of [Equation 2] 　{R, G, B }　The second component of the vector, and the X axis is the time axis, i.e. 1/frame, 2/frame… . or represents the frame number.

FIG. 4G is a diagram 470 showing the result of the autocorrelation function of FIG. 4F.

The Y-axis of FIG. 4g is the autocorrelation function result 　R(y, y), LDA output, X　 is the time axis, that is, 1/frame, 2/frame... . or represents the frame number.

As shown in the drawing described above, after serializing the filtered color information of red, green, and blue as a time function, the heart rate can be estimated using, for example, LDA (Linear Discrimination Analysis) every 5 seconds as a window for a certain time. can In particular, the heart rate cycle may be estimated by separating and extracting signals in which respective color information has similar or different values over time.

In addition, a signal representing heart rate information can be obtained directly from the peak-to-peak distance on the time axis, or the extracted signal can be easily found from the first peak value using an autocorrelation function.

5 is a view 500 for explaining a method for measuring a heartbeat using a color image according to an embodiment.

The heartbeat measurement method using a color image according to an embodiment may store a frame image from the color image (step 501).

A color image is a moving picture obtained from a color camera. If the frame rate is 15 fps or higher and the resolution is 640 x 480 pixels, heart rate measurement is possible.

Next, the heartbeat measurement method using a color image according to an embodiment may select a heartbeat measurement region of interest (ROI). Also, it is possible to process color extraction for the selected region of interest and perform region tracking for the color extracted portion (step 502).

Also, in the heartbeat measurement method using a color image according to an embodiment, the brightness may be corrected after separating red, green, and blue colors for the region-tracked portion (step 503).

The heartbeat measurement method using a color image according to an embodiment extracts a region of interest (ROI) from a color image signal of a camera, calculates a brightness signal, measures the values of individual red, green, and blue signals, can be converted into a signal.

For this, [Equation 1] described above may be used.

The appropriate brightness can be defined as a normalized value of 0.65 to 0.75 to compensate for the red, green, and blue values.

Next, the heartbeat measurement method using a color image according to an embodiment may perform high-pass filtering through 2D high-pass filtering (step 504).

In addition, the heartbeat measurement method using a color image according to an embodiment may perform time domain signal conversion and 5-10 second signal processing window processing (step 505), and perform LDA and autocorrelation function processing (step 506). .

The LDA method has some advantages over PCA or ICA using a simple Eigen vector direction because the LDA method calculates the intertidal solution as in [Equation 2] described above by considering the signal itself and the difference in the average value and the covariance between the signals, respectively. have.

In other words, compared to the sum of the signals, it is advantageous over PCA or ICA because W that maximizes the variance between signals is found.

6 is a diagram 600 for explaining an N-frame heartbeat extraction image sequence window signal.

Reference numeral 601 corresponds to an image sequence removed after signal processing, reference numeral 602 corresponds to an N-frame heartbeat extracted image sequence window signal, and reference numeral 603 corresponds to an n-frame update image sequence.

As shown in FIG. 6 , the heart rate extraction system using a video image outputs heart rate information of n images/m fps. For example, 15 image updates / 30 fps = every 0.5 seconds, heart rate information can be output in real time.

As a result, by using the present invention, it is possible to provide a technology for measuring the heart rate in a non-contact manner using continuous image information of a color image.

In addition, using the present invention, in order to extract a heartbeat, a minute color change between individual color information (red, green, blue) constituting a color image is used, so it is converted into an image of normalized brightness using brightness information of the image Thus, by estimating the heartbeat, errors due to brightness can be minimized. In addition, since it can have sufficient resolution, it is possible to measure heartbeats from adults to infants even with general color video information.

The device described above may be implemented as a hardware component, a software component, and/or a combination of the hardware component and the software component. For example, devices and components described in the embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA). , a microprocessor, or any other device capable of executing and responding to instructions, may be implemented using one or more general purpose computers or special purpose computers. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to execution of the software. For convenience of understanding, although one processing device is sometimes described as being used, one of ordinary skill in the art will recognize that the processing device includes a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that may include For example, the processing device may include a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as parallel processors.

The software may comprise a computer program, code, instructions, or a combination of one or more thereof, which configures a processing device to operate as desired or is independently or collectively processed You can command the device. The software and/or data may be any kind of machine, component, physical device, virtual equipment, computer storage medium or device, to be interpreted by or to provide instructions or data to the processing device. , or may be permanently or temporarily embody in a transmitted signal wave. The software may be distributed over networked computer systems and stored or executed in a distributed manner. Software and data may be stored in one or more computer-readable recording media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the embodiment, or may be known and available to those skilled in the art of computer software. Examples of the computer-readable recording medium include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic such as floppy disks. - includes magneto-optical media, and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

Although the embodiments have been described with reference to the limited drawings as described above, various modifications and variations are possible by those skilled in the art from the above description. For example, the described techniques are performed in a different order than the described method, and/or the described components of the system, structure, apparatus, circuit, etc. are combined or combined in a different form than the described method, or other components Or substituted or substituted by equivalents may achieve an appropriate result.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

Claims (5)

a collection unit for collecting color image signals;
a region of interest extractor for extracting a region of interest from the collected color image signals;
For the extracted color image of the region of interest, a process of converting the extracted brightness signal into a normalized brightness signal using color changes for each of the color information consisting of red, green, and blue, interpolation considering Bayer Pattern A heartbeat estimator for estimating a heartbeat by applying at least one of a preprocessing process using value removal and filtering that considers only the actual pixel value, and a Linear Discrimination Analysis (LDA) process in relation to the color information.
A heartbeat measuring device using a color image comprising a. According to claim 1,
Further comprising a compensator for calculating the brightness signal for the extracted region of interest (ROI), measuring individual red, green, and blue values, and converting it into a normalized brightness signal, and compensating for individual red, green, and blue values A heartbeat measurement device using color images. According to claim 1,
A preprocessor that preprocesses at least one of the collected color image or the extracted region of interest using a 2D high pass filter
further comprising,
The 2D high pass filter considers the Bayer pattern of the color image sensor, and the corresponding color measurement pixel is
A color image in which the value is set as a positive coefficient (+C _ij ), a negative coefficient (-C _ij ) in the case of an interpolation value, and a weight value of a small value (+δ) to the value of the nearest pixel position. heart rate measuring device.
is a heartbeat measurement device using color images. According to claim 1,
The heart rate estimating unit,
A heartbeat measuring apparatus using a color image for estimating a heart rate by applying LDA (Linear Discrimination Analysis) that considers the difference between the average values of the colors for each of the color information. 5. The method of claim 4,
The heartbeat estimation unit according to an embodiment,
After serializing each of the color information as a function of time, the LDA (Linear Discrimination Analysis) is applied to each window of a predetermined time, and signals having similar or different values according to time to the respective color information to which the LDA is applied are separated and extracted. A heartbeat measuring device using color images to estimate the heartbeat cycle.

Images