KR20120045526A - Apparatus and method for eliminating noise - Google Patents

Abstract

PURPOSE: An apparatus and method for removing noises are provided to minimize the distortion of an electrocardiogram signal by estimating a change noise of a base line. CONSTITUTION: A noise removing apparatus includes a detection unit(110) and a control unit(120). The detection unit detects a base signal, determines a size of a structuring element to which a morphology operation is applied and detects the maximum sampling interval. The control unit removes detects an upward peak signal area and a downward peak signal area from an original signal by removing the base signal from the original signal.

Description

Noise canceller and method for canceling noise {Apparatus and method for eliminating noise}

The present invention relates to a noise canceling apparatus and a noise canceling method, and more particularly, to a noise canceling apparatus and a noise canceling method for removing noise by using morphology calculation.

The ECG signal is an electrical record of the heart contraction and relaxation movements caused by the microcurrent generated in the sinus node on the surface of the body and provides important information in diagnosing heart disease of the patient because it provides various information related to the heart activity.

1 is a diagram illustrating an example of a general ECG signal. As shown in FIG. 1, an electrocardiogram signal is a P wave associated with atrial depolarization (contraction), a Q wave representing a low point of the waveform, an R wave representing a high point of the waveform, a low S wave, and a T wave associated with ventricular repolarization. It is composed. The P waves, the QRS-complex waves, and the T waves have various clinical meanings as important waveforms representing the electrical and physiological characteristics of the heart. Easy to get

Particularly, the noise fluctuation of the signal line caused by the breathing of the patient is a low frequency component of less than 1 Hz and has a frequency band similar to that of the ST-segment, which is small in size and low in frequency. Since ST-segment is a signal component that must be measured very accurately for the diagnosis of heart disease such as myocardial infarction or myocardial ischemia, elimination of baseline fluctuation noise signals is essential for the smooth diagnosis of heart disease in patients, where distortion of ST-segment Minimization must be considered.

Conventional methods for removing baseline fluctuation noise include the use of a highpass filter and an adaptive filter of a standard filter. However, in the case of the standard filter, the noise beyond the cutoff frequency band cannot be removed. If the cutoff frequency band is widened to remove the noise, the ECG signal may be distorted. Therefore, the diagnosis parameter may not be accurately extracted. In addition, although the adaptive filter shows relatively good performance, it is difficult to select a reference signal that is an objective reference for normal and abnormal signals. In other words, when an incorrect reference signal is used as an input, it may not remove noise at all or may even cause distortion of the original signal.

Accordingly, a method for minimizing distortion of the original signal and removing the baseline variation noise signal is required.

The present invention has been made to solve the above problems, an object of the present invention, by applying a morphology operation to the original signal containing the baseline variation noise multiple times, to estimate the baseline variation noise, and remove the noise The present invention provides a device and a method for removing noise accordingly.

According to the present invention for achieving the above object, a method for removing noise of a signal is performed by performing a first-order morphology operation to remove a first directional peak on the basis of the magnitude of an input signal, which is the opposite direction of the first directional peak. Performing a second order morphology operation to remove a second directional peak, detecting a base signal, and removing the base signal component from the input signal to detect a component to be measured.

The detecting of the base signal may include detecting a maximum sampling index interval among sampling index intervals of the first direction peak and the second direction peak, and adding a unit sampling index to the maximum sampling index interval. Determining a first interval, performing a first order morphology operation on the first direction of the input signal using the first structural element having the first interval, and using the maximum sampling index interval. Determining a second interval and performing a second morphology operation using a second structural element having the second interval with respect to a second direction of a result signal of the first morphology operation. .

Wherein the first morphology operation comprises (a1) applying the first structural element along a sampling axis with respect to the first direction of the input signal, (b1) within a first interval of the applied first structural element Detecting one of the highest point and the lowest point of the input signal value; (c1) moving the first structural element along the sampling axis and repeating steps (a1) and (b1), wherein the detected value And detecting the signal waveform by connecting them.

Wherein the second morphology operation comprises (a2) applying the second structural element along a sampling axis with respect to the second direction of the resultant signal, (b2) within a second interval of the applied second structural element Detecting the other one of the highest and lowest points of the input signal value, (c2) moving the second structural element along the sampling axis and repeating the steps (a2) and (b2), wherein the detected Concatenating the values to detect the signal waveform.

On the other hand, the input signal is represented by a signal waveform on the secondary domain whose time is the sampling axis and whose magnitude is the vertical axis, and the first directional peak protrudes in one of the upper and lower directions on the vertical axis. The second direction peak may be a peak protruding in the other direction of the upper side and the lower side on the vertical axis.

On the other hand, according to another embodiment of the present invention, the noise removing device for removing the noise of the signal performs a first order morphology operation to remove the first direction peak based on the magnitude of the input signal, the first direction peak A detector for detecting a base signal by performing a second order morphology operation to remove the second direction peak opposite to the direction of the signal; and a controller for removing the base signal component from the input signal and detecting the measurement target component. .

The display apparatus may further include a storage configured to store a result of performing the first morphology operation and the second morphology operation, and a display configured to display a signal from which the base signal component is removed from the input signal. Detecting a maximum sampling index interval among sampling index intervals of a first direction peak and a second direction peak, adding a unit sampling index to the maximum sampling index interval, and determining a first interval; Performing a first order morphology operation on the first direction of the input signal using a first structural element having a value, determining a second time interval using the maximum sampling index interval, and performing the first morphology operation For a second direction of the resultant signal, a second order morphology operation using the second structural element having the second spacing is to be performed. Can be.

Wherein the first order morphology operation comprises (a1) applying the first structural element along a sampling axis with respect to the first direction of the input signal, (b1) within a first interval of the applied first structural element Detecting one of the highest point and the lowest point of the input signal value; (c1) moving the first structural element along the sampling axis and repeating steps (a1) and (b1), wherein the detected value And detecting the signal waveform by connecting them.

Wherein the second order morphology operation comprises (a2) applying the second structural element along a sampling axis with respect to the second direction of the resultant signal, (b2) within a second interval of the applied second structural element Detecting the other one of the highest and lowest points of the input signal value, (c2) moving the second structural element along the sampling axis and repeating the steps (a2) and (b2), wherein the detected Concatenating the values to detect the signal waveform.

On the other hand, the input signal is represented by a signal waveform on the secondary domain whose time is the sampling axis and whose magnitude is the vertical axis, and the first directional peak protrudes in one of the upper and lower directions on the vertical axis. The second direction peak may be a peak protruding in the other direction of the upper side and the lower side on the vertical axis.

The noise removing device and the noise removing method according to the embodiment of the present invention can estimate the baseline fluctuation noise in the original signal to remove the baseline fluctuation noise while minimizing the distortion of the far electrocardiogram signal.

1 is a diagram illustrating an example of a general electrocardiogram signal;
2a to 2d are diagrams showing examples of basic operations constituting a general morphology;
3 is a block diagram illustrating a configuration of an apparatus for removing noise according to an embodiment of the present invention;
4A and 4B are diagrams for describing an operation of the detector 110, according to an exemplary embodiment.
5 is a block diagram showing a configuration of an apparatus for removing noise according to another embodiment of the present invention;
6A to 7B are graphs showing simulation results of estimating and removing baseline fluctuation noise having a form of monotone increase in an electrocardiogram signal using a noise removing device according to an embodiment of the present invention;
8A to 9B are graphs showing simulation results of estimating and removing baseline fluctuation noise having a sinusoidal wave form from an electrocardiogram signal using a noise removing device according to an embodiment of the present invention;
10A to 10C are graphs showing simulation results using a noise removing device according to an embodiment of the present invention, for an ECG signal having a baseline variation noise having a difference between a maximum value and a minimum value of about 0.2 for 1600 sample intervals. ,
11A to 11C are graphs showing simulation results using a noise removing device according to an embodiment of the present invention, for an electrocardiogram signal when the baseline fluctuation noise repeats increase and decrease in one cycle, and
12 is a flowchart illustrating a method of removing noise according to an embodiment of the present invention.

Hereinafter, with reference to the drawings will be described the present invention in more detail.

The present invention aims to remove the baseline fluctuation noise from the original signal by applying the morphology operation a plurality of times, and the morphology calculation used in the present invention will be briefly described before explaining the present invention.

Morphology, which is mostly used in image signal processing, is a nonlinear image processing and analysis method that focuses on the geometrical structure of an image, and is a primary operation of erosion. And opening and closing, which are dilation and secondary operations. Many of these linear and nonlinear operators can be constructed with this simple combination of morphologies.

First, the most basic operations in morphology are erosion and expansion operations. Specifically, the expansion operation is an operation of determining a maximum value of each predetermined threshold set of the voice signal image as a value of the corresponding interval, and tends to shrink the input image. In addition, the erosion operation is an operation for determining a minimum value of each predetermined critical section of the voice signal image as a value of the corresponding interval, and is a dual operation of erosion, which is defined as a set complementation of erosion. Can be.

On the other hand, the quadratic operations of morphology are open and closed. Open is an operation that performs the expansion operation after the erosion operation, and shows a smoothing effect. In addition, the closing is an operation that performs an erosion operation after the expansion operation, and exhibits a filling effect.

Such morphology operations can serve to estimate the overall flow of the signal, excluding regions around the local extrema of the signal, and in particular, can estimate the base signal in an ECG signal that includes baseline variation noise. .

Specific examples in which the above-described erosion, expansion, open, and close operations of the morphology are applied are shown in FIGS. 2A to 2D.

Specifically, FIGS. 2A and 2B show results of performing erosion and expansion operations on the original signal f, respectively. Also, FIG. 2C shows an open operation for the original signal f (FIG. 2A) to which the erosion operation has been applied, that is, an open operation for the original signal f. FIG. 2D shows an erosion operation to the original signal f (FIG. 2B) to which the expansion operation has been applied. Shows the result of the applied, that is, the closing operation on the original signal f.

On the other hand, the noise canceller according to an embodiment of the present invention estimates the base signal from the input signal (hereinafter referred to as the original signal) by performing a plurality of morphological operations on the original signal including the baseline variable noise signal And remove it.

As described above, the morphology operation removes an upward peak signal region having a local maxima component and a downward peak signal region having a local minima component based on the magnitude of the original signal. Means an operation that can estimate the base signal. Here, the upward peak signal region means a signal protruding upward from the baseband based on the magnitude of the original signal, and the downward peak signal region means a signal protruding downward from the baseband based on the magnitude of the original signal. can do.

In particular, the noise canceller can be applied to electrocardiogram signals including baseline fluctuation noise signals generated by the patient's breathing, as well as to the effects of 60Hz modulation waves, wideband white noise, and peripheral devices on power lines. It can also be applied to ECG signals that include muscle noise caused by the movement of the patient.

In addition, in performing the morphology calculation, the noise canceling apparatus may use the sampled original signal by sampling the original signal at a constant sampling frequency. Therefore, when the original signal is displayed as a signal waveform on the secondary domain, the horizontal axis may represent the sampling axis and the vertical axis may represent the magnitude, and the horizontal axis of the upstream and downstream peak signal regions of the original signal may be represented by sampling index intervals.

As such, the apparatus for removing noise according to an embodiment of the present invention includes a detector 110 and a controller 120 as shown in FIG. 3.

The detector 110 performs a morphology operation on the original signal to estimate the base signal of the original signal. To this end, the detector 110 may determine the size of a structural element to which the morphology operation is applied. Specifically, the maximum sampling index interval among the sampling index intervals of the upstream and downstream peak signal regions in the baseband of the original signal is detected, and the unit sampling index is added to the detected maximum sampling index interval to add a structural element. The size of can be determined.

Herein, the structural element refers to a determination unit for determining an upstream peak signal and a downstream peak signal region to be removed from an original signal in performing a morphology operation, and the size of the structural element may be defined as a sampling index interval.

As an example, if a morphology operation is performed on a raw signal by applying a structural element having a size of +1, the up and down peak signals having a sampling index interval of 0 to μ in the base band of the original signals are removed. The upward and downward peak signals having a sampling index interval of μ + 1 or more are not eliminated. In other words, in order to remove a signal region having a sampling index interval of 0 to μ, a unit sampling index may be added to the maximum sampling index interval among the signal regions to be removed to determine the size of the structural element.

Accordingly, by appropriately setting the size of the structural element of the morphology operation, it is possible to determine the areas of the upstream and downstream peak signals to be removed from the original signal.

In addition, the detector 110 applies a determined structural element to perform a morphology operation on the original signal. Specifically, the erosion operation may be performed during the morphology operation to remove the upward peak signal region, the expansion operation may be performed to remove the downward peak signal region, and the determined structural element may be applied according to the sampling axis. Can be.

Specifically, the detection unit 110 moves the determined structural element along the original signal when performing the erosion operation, detects the lowest point of the original signal value within the magnitude of the structural element, and connects the detected values to detect a signal waveform. do. In addition, during the expansion operation, the determined structural element is moved according to the original signal, the highest point of the original signal value is detected within the magnitude of the structural element, and the detected values are connected to generate a signal waveform.

As an example, it is assumed that the detector 110 sets the size of the structural element to μ + 1 in order to remove the upward peak signal and the downward peak signal having a sampling index interval of 0 to μ. Thereafter, when the detector 110 performs the erosion operation with the size of the structural element determined as μ + 1, the sampling index is less than μ + 1 in the base band of the original signal, that is, the sampling index interval is 0 to μ. The upward peak signal region is replaced by a value similar to the surrounding background signal and removed. However, for the downward peak signal region, the detector 110 detects the lowest point of the downward peak signal region within μ + 1 and connects the detected values to generate a signal waveform. The generated signal waveform has a sampling index interval (2 μ) corresponding to twice the maximum sampling index interval (corresponding to μ in this example) among the sampling index intervals of the up-peak and down-peak signal regions to be removed.

The same principle can be applied to the case where the detector 110 performs the expansion operation. However, as opposed to the erosion operation, the downward peak signal region is removed during the expansion operation, and the upward peak signal region is μ + 1. The highest point of the upward peak signal region is detected within the signal waveform, and a signal waveform to which the detected peak is connected is generated.

Meanwhile, the detector 110 may re-determine the size of the structural element with respect to the original signal on which the morphology operation is performed once. As described above, the sampling index interval that the signal on which the morphology operation is performed once in the base band is determined by the maximum sampling index interval of the original signal. That is, the up-peak or down-peak signal region of the signal on which the morphology operation is performed once has an interval corresponding to twice the maximum sampling index interval among the sampling index intervals of the up-peak and down-peak signal regions of the original signal.

Therefore, the detector 110 may re-determine the size of the structural element by adding the unit sampling index to a value doubled to the maximum sampling index interval of the original signal.

In addition, the detector 110 may re-perform the morphology operation once again on the signal in which the morphology operation is performed once on the original signal by applying the size of the reconstructed structural element. This is to remove the up-peak signal or the down-peak signal present in the signal in which the morphology operation is performed once. Since the method of applying the morphology operation has been described in detail above, redundant description thereof will be omitted.

In the previous example, when an erosion operation is performed on the original signal, the upward peak signal region is removed, leaving a downward peak signal having a sampling index interval of 2 mu. To eliminate this, the detector 110 re-determines the size of the structural element at 2μ + 1 and performs an expansion operation on the size of the structural element at 2μ + 1, so that the downward peak present in the signal in which the morphology operation is performed once The signal area can be removed.

As such, the detection unit 110 re-determines the size of the structural element with respect to the signal on which the morphology operation is performed once, and re-executes the morphology operation by applying the re-determined structural element, thereby generating the upper peak and the lower peak signal regions in the original signal. You can remove them all. Accordingly, it is possible to estimate a signal from which the up-peak and down-peak signal regions are removed from the original signal, that is, the base signal of the original signal.

 The controller 120 may remove a signal in which the morphology operation is performed twice from the original signal. Specifically, since the signal in which the morphology operation is performed twice is a signal in which both the upstream peak signal and the downstream peak signal region have been removed from the original signal, that is, the base signal, the control unit 120 calculates arithmetic difference from the base signal from the original signal. By using this method, it is possible to finally detect the upward peak and downward peak signal regions in the original signal.

On the other hand, in the above-described embodiment, it is assumed that the original signal refers to an input signal input to the noise canceller, but this includes not only an input signal but also a signal stored in the noise canceller or a signal measured directly. .

Although the apparatus for removing noise according to an embodiment of the present invention may be applied to all signals including the baseline variation noise signal, the ECG signal including the baseline variation noise signal is taken as an example and is shown in FIGS. 4A and 4B. With reference to, the operation of the detector 110 will be described.

The signal shown by the solid line in FIG. 4A is an electrocardiogram signal, and the upstream peak signal region corresponds to the P, R, and T wave signal regions, and the downside peak signal region corresponds to the Q wave and S wave signal regions.

Since the sampling index interval of the QRS-complex in the base band among the P-wave, QRS-complex, and T-wave signal regions in the ECG signal shown in FIG. 4A is μ, the detector 110 is larger than μ, which is the maximum sampling index interval. You can determine the size of the structural element with μ + 1.

Subsequently, when the erosion operation is performed on the structural element of μ + 1 to remove the P, R, and T waves, which are the upstream peak signals, the upstream peak signal having a spacing of 0 to μ is removed, and Since the structural elements move along the sampling axis and detect the lowest point within the μ + 1 interval to generate a signal connecting them, the Q peak and S waves, which are the downward peak signal regions, remain and are sampled, as shown by the dotted line shown in FIG. 4A. The index interval is 2μ.

Accordingly, in order to remove the Q and S waves, which are the downward peak signal areas after the erosion operation, the detector 110 resets the size of the structural element to 2μ + 1 and applies the reset structural element to perform the expansion operation. Thus, the Q and S waves, which are the downward peak signal region, can be removed.

Although it is assumed in the above-described embodiment that the detection unit 110 performs an erosion operation and performs an expansion operation, the order may be changed. That is, the detector 110 may first perform an expansion operation to remove the downward peak signal first, and then perform an erosion operation to remove the remaining upward peak signal. Even in this case, the detector 110 may determine the size of the structural element at μ + 1 for the expansion operation and re-determine the size of the structural element at 2μ + 1 for the erosion operation.

Accordingly, it is possible to estimate the base signal by removing both the upstream peak and the downward peak signal regions from the ECG signal.

5 is a block diagram illustrating a configuration of an apparatus for removing noise according to another embodiment of the present invention. In the noise canceling apparatus of FIG. 5, a redundant description is omitted for a configuration that performs the same operation as that of FIG. 3.

The storage unit 230 stores a signal. Specifically, the signal input to the present noise canceller, the signal obtained by performing a plurality of morphology operations on the original signal by the detector 210, and the up-peak and down-peak signals of the original signal detected by the controller 120. Can store information about

In addition, the storage 230 may transmit the stored signal to the detector 210 and the controller 220.

The display unit 240 performs an operation of displaying a signal processed by the noise removing device. Specifically, when the signal in which the morphology operation is performed twice from the original signal is removed by the controller 220, the controller 220 may display the removed signal.

In addition, the display unit 240 may display a signal having a predetermined magnitude in displaying a signal from which a morphology operation is performed twice from an original signal.

For example, since the signal in which the morphology operation is performed twice is a base signal estimated from the ECG signal, when the signal in which the morphology operation is performed twice in the ECG signal is removed, the signal region except for the QRS-complex, P wave, and T wave is The signal magnitude is approximated to zero. Accordingly, the display unit 240 may display the signal more clearly by applying a DC-like signal through a bias value to the signal from which the morphology calculation is performed twice from the ECG signal.

The operation of the display unit 240 may be automatically performed or may be performed by user's setting.

Hereinafter, referring to FIGS. 6A to 11C, a simulation result of a noise removing device according to an exemplary embodiment of the present invention will be described using an ECG signal including a baseline variable noise signal as an example.

Here, the ECG signal has a sampling frequency of 360 Hz, and the ECG signal of the MIT / BIH database measured for about 10 minutes from each examiner was used, and the experimental PC is a specification of a 3.00GHz Core2 Duo CPU and 3GByte memory, and Visual C ++. 6.0 compiler was used. In addition, time is shown by sampling at a period of 0.0028 seconds, represented by a sampling index on each figure.

6A to 7B are graphs showing simulation results of estimating and removing baseline variation noise having a form of monotone increase in an electrocardiogram signal by using a noise removing device according to an embodiment of the present invention.

6A to 6C illustrate an example of generating and applying an artificial baseline variation noise signal using a logarithmic function in the form of monotone increase. Since the magnitude of the ECG source signal shown in FIG. 6A is in the range of about -0.5 to 1, the difference between the minimum value and the maximum value is about 1.5, and the artificial baseline variation noise signal of FIG. 6B shows monotonic increase in the range of about 0.2 to 1.2. Has characteristics. The artificial baseline variable noise applying signal derived through the arithmetic sum of FIGS. 6A and 6B is the same as that of FIG. 6C, and has the meaning of applying the off-set signal of FIG. 6B to FIG. 6A.

7A and 7B are graphs showing simulation results obtained for the signal of FIG. 6C using a noise removing device according to an embodiment of the present invention.

FIG. 7A illustrates a baseline fluctuation noise signal estimated by a noise removing apparatus according to an exemplary embodiment. FIG. 7B illustrates a signal from which an estimated baseline fluctuation noise signal is removed from an EKG signal. FIG. The baseline variation noise canceling signal may be represented as shown in FIG. 7B using the arithmetic difference of FIGS. 6C and 7A.

6b and 7a, it can be seen that the noise canceller according to an embodiment of the present invention can estimate the baseline fluctuation noise signal excellently, and the difference between the two signals in some intervals is the ECG source signal. It is caused by minute baseline fluctuation noise, power supply noise, and the like of FIG. 6A and has a level of exclusion. In addition, the baseline fluctuation noise is removed, Figure 7b can be confirmed that the baseline fluctuation noise is effectively removed while maintaining the value of about 1.5, the original signal amplitude width of Figure 6a.

8A to 9B are graphs showing simulation results of estimating and removing baseline fluctuation noise having a sinusoidal wave form from an ECG signal using a noise removing device according to an exemplary embodiment of the present invention.

The signal shown in FIG. 8A is an example of an ECG signal. Since the magnitude is in the range of about -0.5 to 1, the difference between the minimum value and the maximum value is about 1.5. In addition, the signal shown in FIG. 8B is an artificial baseline variation noise signal and has a frequency of 10 Hz in the range of -0.8 to 0.8. Figure 8c shows an electrocardiogram signal with baseline fluctuation noise derived from the arithmetic sum of Figures 8a and 8b.

FIG. 9 is a graph illustrating simulation results obtained for an electrocardiogram signal including baseline variation noise illustrated in FIG. 8C by using a noise removing device according to an exemplary embodiment. FIG. 9A shows the estimated baseline variation noise signal, and FIG. 9B shows the result of setting 0 to bias.

As shown in FIG. 7A and FIG. 7B, it can be seen from FIG. 9A that the proposed method can be excellently estimated even when the sinusoidal baseline fluctuation noise is introduced. In addition, it can be seen that only the baseline fluctuation noise of the sine wave characteristic was removed while maintaining the signal amplitude width of the original signal in FIG. 9B which is the baseline fluctuation noise removing signal.

10A to 10C are graphs showing simulation results using a noise removing device according to an embodiment of the present invention, for an ECG signal having a baseline variation noise having a difference between a maximum value and a minimum value of about 0.2 for 1600 sample intervals. to be.

FIG. 10A illustrates an electrocardiogram signal having a baseline variation noise having a difference between a maximum value and a minimum value of about 0.2 for 1600 sample intervals, and FIG. 10B is an estimated baseline variation noise signal. You can see that it is similar to the width. In addition, FIG. 10C is a graph showing the ECG signal from which the baseline variation noise is removed, and thus it can be seen that the baseline variation noise is effectively removed. In particular, even when the original signal does not have the local minimum value of the Q and S waves unlike the normal ECG signal, it can be confirmed from FIG. 10C that the signal distortion of the Q and S waves has not occurred on the resultant signal. .

11A to 11C are graphs showing simulation results using a noise removing device according to an embodiment of the present invention, for an electrocardiogram signal when the baseline variation noise repeats increase and decrease in one cycle.

FIG. 11A illustrates an ECG signal in which the baseline fluctuation noise repeatedly increases or decreases in one cycle with respect to an ECG signal cycle consisting of P waves, QRS-complex, and T waves, and in this case, sinusoidal baseline fluctuations. It can be regarded as a noise introduced form. 11B is an estimated baseline variation noise signal, and it can be seen that baseline variation noise in the form of repeated increase and decrease is inferred. In addition, FIG. 11C illustrates a signal in which the baseline fluctuation noise is removed from the signal shown in FIG. 11A, and it can be confirmed that the baseline fluctuation noise in the form of repeated increase and decrease in one cycle can be effectively removed.

12 is a flowchart illustrating a method of removing noise according to an embodiment of the present invention.

Referring to FIG. 12, a morphology operation is performed on the original signal in order to remove the up-peak signal or the down-peak signal based on the magnitude of the original signal (S310). To this end, a maximum sampling index is detected among the sampling index intervals of the up-peak signal or the down-peak signal of the original signal, the unit sampling index is added to the detected maximum sampling index, and the size of the structural element is set to perform a morphology operation. Can be.

Thereafter, the morphology calculation is performed again to remove the upper peak signal or the lower peak signal present after the morphology calculation. When the morphology operation is performed again, since both the upper peak signal and the lower peak signal are removed from the original signal, the base signal is detected from the original signal through this (S320).

Thereafter, the detected base signal can be removed from the original signal to detect the component to be measured (S330).

On the other hand, while the above has been shown and described with respect to the preferred embodiments of the present invention, the present invention is not limited to the specific embodiments described above, the technical field to which the invention belongs without departing from the spirit of the invention claimed in the claims Of course, various modifications can be made by those skilled in the art, and these modifications should not be individually understood from the technical spirit or the prospect of the present invention.

110, 210: detector 120, 220: controller
230: storage unit 240: display unit

Claims (10)

In the noise reduction method of the signal,
Performing a first order morphology operation to remove the first direction peak based on the magnitude of the input signal, and perform a second order morphology operation to remove the second direction peak opposite to the first direction peak, Detecting the base signal; And
And removing the base signal component from the input signal to detect the component to be measured. The method of claim 1,
Detecting the base signal,
Detecting a maximum sampling index interval among sampling index intervals of the first direction peak and the second direction peak;
Determining a first interval by adding a unit sampling index to the maximum sampling index interval;
Performing a first order morphology operation on a first direction of the input signal using the first structural element having the first spacing;
Determining a second interval using the maximum sampling index interval; And
And performing a second morphology operation using the second structural element having the second interval with respect to a second direction of the result signal of the first morphology operation. The method of claim 2,
The first morphology operation is,
(a1) applying the first structural element along a sampling axis with respect to the first direction of the input signal;
(b1) detecting one of a highest point and a lowest point of the input signal value within a first interval of the applied first structural element;
(c1) detecting the signal waveform by connecting the detected values while moving the first structural element along the sampling axis and repeating steps (a1) and (b1). Noise Reduction Method. The method of claim 3,
The second morphology operation,
(a2) applying said second structural element along a sampling axis with respect to said second direction of said result signal;
(b2) detecting, within the second interval of the applied second structural element, the other of the highest and lowest points of the input signal value;
(c2) detecting the signal waveform by connecting the detected values while moving the second structural element along the sampling axis and repeating steps (a2) and (b2). Noise Reduction Method. 4. The method according to any one of claims 1 to 3,
The input signal is,
Represented by a signal waveform on the secondary domain with the sampling index on the horizontal axis and the magnitude on the vertical axis,
The first direction peak is a peak protruding in any one of an upward direction and a downward direction on the vertical axis,
And the second direction peak is a peak protruding in the other direction of the upper and lower directions on the vertical axis. A noise canceling device for removing noise in a signal,
Performing a first order morphology operation to remove the first direction peak based on the magnitude of the input signal, and perform a second order morphology operation to remove the second direction peak opposite to the first direction peak, A detector for detecting a base signal;
And a controller which removes the base signal component from the input signal and detects a component to be measured. The method of claim 6,
A storage unit to store a result of performing the first and second morphology operations; And
And a display unit configured to display a signal from which the base signal component is removed from the input signal.
Wherein:
Detecting a maximum sampling index interval among sampling index intervals of the first and second directional peaks, and adding a unit sampling index to the maximum sampling index interval to determine a first interval; A first order morphology operation is performed on the first direction of the input signal using a first structural element having a spacing, a second distance is determined using the maximum sampling index interval, and the first morphology calculation is performed. And a second order morphology operation using the second structural elements having the second spacing for a second direction of the resulting signal. The method of claim 7, wherein
The first morphology operation is,
(a1) applying the first structural element along a sampling axis with respect to the first direction of the input signal;
(b1) detecting one of a highest point and a lowest point of the input signal value within a first interval of the applied first structural element;
(c1) detecting the signal waveform by connecting the detected values while moving the first structural element along the sampling axis and repeating steps (a1) and (b1). Noise Canceling Device. The method of claim 8,
The second morphology operation is,
(a2) applying said second structural element along a sampling axis with respect to said second direction of said result signal;
(b2) detecting, within the second interval of the applied second structural element, the other of the highest and lowest points of the input signal value;
(c2) detecting the signal waveform by connecting the detected values while moving the second structural element along the sampling axis and repeating steps (a2) and (b2). Noise Canceling Device. 9. The method according to any one of claims 6 to 8,
The input signal is,
Represented by a signal waveform on the secondary domain with the sampling index on the horizontal axis and the magnitude on the vertical axis,
The first direction peak is a peak protruding in any one of an upward direction and a downward direction on the vertical axis,
And the second direction peak is a peak protruding in the other direction of the upper and lower directions on the vertical axis.

Images