JP4661115B2 - Signal processing apparatus, signal processing method, and mechanical apparatus - Google Patents

Description

  The present invention relates to a signal processing apparatus and signal processing method for acquiring a bioelectric signal from a living body surface, and a mechanical apparatus, and more particularly to a signal processing apparatus and signal processing for acquiring a bioelectric signal resulting from voluntary movement and extracting a feature quantity. The present invention relates to a method and a mechanical device controlled based on a signal extracted from a bioelectric signal resulting from voluntary movement of the living body.

  Among bioelectric signals that can be measured non-invasively by attaching electrodes to the surface of the living body, myoelectric potential and ocular potential are examples of signals that reflect voluntary actions. In particular, attempts have been made for a long time to apply a myoelectric potential waveform pattern as a control signal for a medical instrument such as a prosthetic hand (robot hand). In recent years, not only medical applications but also applications as information devices and other input devices have been studied (see Patent Document 1 and Patent Document 2). Information contained in the myoelectric signal includes an electromyogram (EMG) obtained by amplifying and filtering the detected myoelectric potential, and an integrated electromyogram (IEMG) obtained by rectifying and smoothing the EMG. Integrated Electro Myo-Gram) and EMG frequency components. Further, the frequency components of EMG and IEMG are weak to the influence of sensor contact resistance and noise, and the detection error becomes large. Therefore, by utilizing the temporal change of spatial myoelectric potential distribution or spatial myoelectric potential distribution, A technique for improving the identification accuracy of a control signal to a control object associated with body motion or body motion is disclosed (see Patent Document 2).

JP-A-7-248873 JP-A-8-036459 JP 2002-287869 A

  The technique described in Patent Document 1 is a technique for controlling a control target based on the motion and force of each part of the body detected as a myoelectric signal, and is complicated in the combination of finger motion and hand gripping force. By assigning the command, even a user who is not familiar with the keyboard operation can control the controlled object regardless of the keyboard. Patent Document 2 discloses a pointing device that detects a myoelectric potential signal, converts it into position information, and displays a cursor.

  However, the myoelectric potential (surface myoelectric potential) measured by electrodes on the surface of a living body is detected as a superposition of action potentials by a plurality of motor units, and the action potentials of not only a single muscle but also a plurality of muscles are combined. Detected. It is also known that the myoelectric potential waveform varies greatly depending on the electrode mounting position, and in the bipolar derivation in which the electrode is attached in the direction along the muscle fiber across the innervation zone, the myoelectric potential component is attenuated by interference. . For this reason, the myoelectric potential measurement using the skin surface has a problem that the reproducibility of the waveform is generally poor even in the measurement for the same individual. Since there are individual differences in the muscle running and the position of the innervation zone, it is more difficult to associate the body movement of the unspecified person with the myoelectric potential waveform. Further, in the technique of Patent Document 2, the burden on the subject is heavy, such as the user needs to train physical movements for accurate detection.

  In addition, when measuring bioelectric signals by attaching electrodes to the surface of a living body, excluding special noise such as a shielded room, it is inevitable that external noise such as commercial power will be mixed, and skin caused by perspiration Noise derived from a change in resistance or a change in contact resistance due to a change in the degree of adhesion of the electrode to the skin may be mixed.

  Thus, the frequency components of EMG and IEMG are vulnerable to individual differences in the movement of the measurement site, changes in the contact resistance of the sensor, the influence of noise, etc., and the detection error increases each time. . Since the detection error affects the identification of the myoelectric potential signal, in order to use the myoelectric potential signal as a control signal to be controlled, it is important to improve the identification rate of the myoelectric potential signal.

  The technique described in Patent Document 3 was made for the purpose of improving the recognition rate of the myoelectric potential signal. However, a control signal associated with a predetermined reference pattern (reference myoelectric potential distribution) is obtained. In order to output, it is necessary to calculate the correlation between the spatial myoelectric potential distribution calculated from the sensor results of the plurality of myoelectric potential detection sensors and the reference pattern.

  There are other methods to match the muscle movement to the myoelectric waveform by identifying the difference in the frequency distribution of the myoelectric waveform or identifying the amplitude pattern of the myoelectric potential. Poor sex has led to a worsening of recognition rate.

  Therefore, an object of the present invention is to provide a signal processing apparatus and a signal processing method that can generate a binary pulse signal from a bioelectric signal waveform resulting from voluntary movement and identify a pulse width or a pulse pattern.

In order to achieve the above-described object, a signal processing apparatus according to the present invention includes a bioelectric signal acquisition unit that acquires a bioelectric signal resulting from voluntary movement, and a bioelectric signal acquired by the bioelectric signal acquisition unit. The smoothing means to be converted, the threshold holding means for holding the threshold of the amplitude of the bioelectric signal, and the amplitude threshold held in the threshold holding means and the amplitude of the bioelectric signal acquired in the bioelectric signal acquiring means Comparing means, and pulse signal generating means for extracting, as one pulse signal, a period of continuous bioelectric signals having an amplitude exceeding a threshold value from the comparison result in the comparing means among the bioelectric signals smoothed by the smoothing means , and the pulse signal output means for outputting a pulse signal generated in the pulse signal generating means, the pulse signal generated by the pulse signal generating means The difference between the rise time and the fall time and detects the pulse width, Ru and a pulse pattern detecting means for detecting a pulse interval between adjacent pulses of the pulse signal generated by the pulse signal generating means. Since the signal processing apparatus according to the present invention does not extract the characteristics of the bioelectric signal itself, the reproducibility of the bioelectric signal waveform is not greatly affected by the recognition rate.

  The signal processing apparatus according to the present invention includes a threshold value holding unit that holds a threshold value of the amplitude of the bioelectric signal, a threshold value of the amplitude held in the threshold value holding unit, and the bioelectric signal acquired by the bioelectric signal acquisition unit. And a comparison means for comparing. In this case, the pulse signal generating means extracts a period in which bioelectric signals having an amplitude exceeding the threshold value from the comparison result in the comparing means as one pulse signal.

  You may provide the threshold value setting means which sets the threshold value of a bioelectric signal. In this case, the comparison means compares the amplitude threshold set by the threshold setting means with the bioelectric signal acquired by the bioelectric signal acquisition means, and the pulse signal generation means causes the amplitude exceeding the threshold from the comparison result of the comparison means. A period in which a bioelectric signal having a signal is continuous is extracted as one pulse signal.

  In addition, the pulse width of the pulse signal generated from the bioelectric signal is detected by providing pulse width detection means for detecting the difference between the rise time and the fall time of the pulse signal generated by the pulse signal generation means as the pulse width. A plurality of control signals can be assigned to combinations of pulse signals.

  When the rising time of the first pulse signal is T1, the falling time T2, the rising time of the second pulse signal is T3, and the falling time is T4, (T3-T1), A pulse interval between adjacent pulse signals can be detected by any one of (T3-T2), (T4-T1), and (T4-T2).

  The signal processing apparatus according to the present invention may include a plurality of bioelectric signal acquisition means. At this time, the pulse signal generation means extracts a period in which bioelectric signals having amplitudes exceeding a threshold for each of the plurality of bioelectric signals are continuous as one pulse signal, so that more than the plurality of pulse signals are obtained. Many control signals can be assigned.

In order to achieve the above-described object, a signal processing method according to the present invention includes a bioelectric signal acquisition step for acquiring a bioelectric signal resulting from voluntary movement, and a smoothing of the bioelectric signal acquired in the bioelectric signal acquisition step. And a comparison step for comparing the amplitude threshold held in the threshold holding means for holding the threshold of the amplitude of the bioelectric signal with the amplitude of the bioelectric signal acquired in the bioelectric signal acquisition step of the smoothed bioelectrical signal in the smoothing process, and the pulse signal generating step of extracting a period in which bioelectrical signal having an amplitude that exceeds the threshold value from the comparison result of the comparing step is continuous as a pulse signal, the pulse signal and the pulse signal output step of outputting a pulse signal generated in the generating step, rising time and falling of the pulse signal generated by pulse signal generation step It detects a difference in time as the pulse width, to have a pulse pattern detection step of detecting a pulse interval between adjacent pulses of the pulse signal generated by the pulse signal generation process. In the signal processing method according to the present invention, since the characteristics of the bioelectric signal itself are not extracted, the reproducibility of the bioelectric signal waveform is not greatly affected by the recognition rate.

In addition, the mechanical device according to the present invention includes a bioelectric signal acquisition means for acquiring a bioelectric signal resulting from voluntary movement in a mechanical device whose operation is controlled based on a control signal input from a user, Smoothing means for smoothing the bioelectric signal acquired in the signal acquisition means, threshold holding means for holding the threshold of the amplitude of the bioelectric signal, threshold of amplitude held in the threshold holding means, and bioelectric signal acquisition means Comparison period for comparing the amplitude of the bioelectric signal acquired in step 1 and the bioelectric signal smoothed by the smoothing means are continuous periods of bioelectric signals having an amplitude exceeding a threshold value from the comparison result of the comparison means A pulse signal generation means for extracting a signal as a single pulse signal, a storage means for storing a signal pattern of the pulse signal and a control signal in association with each other, and a pulse And the pulse signal output means for outputting a pulse signal generated in the No. generation means, a difference between the rise time and the fall time of the pulse signal generated by the pulse signal generating means and detects a pulse width, the pulse signal generating means A pulse pattern detection means for detecting a pulse interval between adjacent pulses of the generated pulse signal, and a control signal corresponding to the pulse signal pattern output from the pulse pattern detection means is selected from the storage means, and the control signal is Ru and control means for performing control to change the operation.

  According to the signal processing device and the signal processing method of the present invention, the reproducibility of the signal obtained from the bioelectric signal is improved by generating the pulse signal from the bioelectric signal waveform. Further, when a specific control signal is assigned to a pulse signal generated from a bioelectric signal and applied to device control, the recognition rate of the bioelectric signal is improved by improving the reproducibility of the pulse signal. Accordingly, an appropriate control signal assigned to the pulse signal can be selected. Thereby, a new entertainment use and a new technical use can be created.

  Specific examples of the present invention will be described in detail with reference to the drawings. A signal processing apparatus 1 shown as a specific example of the present invention is a signal processing apparatus 1 that generates a pulse signal by detecting a bioelectric signal resulting from voluntary movement. Bioelectric signals resulting from voluntary movement include myoelectric potential and ocular potential. Hereinafter, an example in which myoelectric potential is a detection target will be described.

  A signal processing apparatus 1 shown in FIG. 1 has a myoelectric potential detection sensor 11 as bioelectric signal acquisition means for detecting a myoelectric signal that is one of bioelectric signals caused by voluntary movement, and the detected myoelectric potential signal. Biological amplifier 12 to be amplified, A / D conversion circuit 13 for converting the amplified analog myoelectric potential signal into a digital signal, a high pass filter 14 for cutting noise, and the myoelectric potential signal are rectified. A full-wave rectifier circuit 15, a low-pass filter (Low Pass Filter) 16 as a smoothing means for smoothing the myoelectric potential signal detected by the myoelectric potential detection sensor 11, and a threshold value of the smoothed myoelectric signal. A pulse generator 17 serving as a pulse signal generator for extracting a period of continuous myoelectric potential signals having an amplitude exceeding 1 as a pulse signal, and an output terminal 1 for outputting a binarized pulse signal 8 and.

  The myoelectric potential detection sensor 11 is a skin surface electrode attached to the skin surface of the subject (user), and is attached to the skin surface such as a hand or an arm, and detects a myoelectric potential signal generated by the body motion of the subject. The detected myoelectric potential signal is sent to the biological amplifier 12. The myoelectric potential detection sensor 11 may be composed of a plurality of skin surface electrodes. In this case, if it is worn at a plurality of locations on the subject's body, it is possible to detect myoelectric potential signals due to a plurality of types of body movements. The myoelectric potential signal changes depending on the activity of the muscle cell, and the amplitude of the detected myoelectric potential signal varies depending on the amount of fat and the strength of the muscle activity.

  The biological amplifier 12 is a signal amplifier and amplifies the myoelectric potential signal from the myoelectric potential detection sensor 11. Since the myoelectric potential signal detected by the myoelectric potential detection sensor 11 is usually about several tens of millivolts, it is necessary to amplify it to a level that allows signal analysis. The biological amplifier 12 sends the amplified myoelectric potential signal to the A / D conversion circuit 13.

  The A / D conversion circuit 13 converts the analog myoelectric potential signal amplified by the biological amplifier 12 into a digital signal, and sends the myoelectric potential signal after A / D conversion to the high-pass filter 14. The waveform of the myoelectric potential signal A / D converted by the A / D conversion circuit 13 is shown in FIG. FIG. 2 shows the amplitude (vertical axis) of the myoelectric potential signal with respect to the period (horizontal axis). The waveform shown in the figure represents a change in muscle action potential (myoelectric potential) generated by muscle activity. The amplitude of the myoelectric potential signal represented on the vertical axis represents the intensity of the muscle activity. Representative ones are indicated by Aa, Ac, Ae, and Af. Further, Ta, Tb, Tc, Td, Te, Tf, and Tg indicate periods of muscle activity. Although not shown in FIG. 2, a waveform having a certain amplitude may be continued for a certain period.

  The high-pass filter 14 removes noise of the DC component of the myoelectric potential signal A / D converted by the A / D conversion circuit 13 and sends the noise-removed myoelectric potential signal to the full-wave rectifier circuit 15. The full-wave rectifier circuit 15 takes the absolute value of the myoelectric potential signal detected by the myoelectric potential sensor 11 and sends it to the low-pass filter 16.

  The low-pass filter 16 is a smoothing circuit that smoothes the rectified myoelectric potential signal. The low-pass filter 16 smoothes the signal from the full-wave rectification circuit 15 and sends it to the pulse generator 17. The waveform of the myoelectric potential signal smoothed by the low-pass filter 16 is shown in FIG. The myoelectric potential waveform shown in FIG. 3 is smoothed by taking the absolute value of the myoelectric potential waveform shown in FIG. 2, and the characteristics of myoactivity appear more clearly. The waveform detected over the period Ta in FIG. 2 corresponds to the peak Sa having the maximum amplitude Asa in FIG. 3, the waveform in the period Tc corresponds to the peak Sc having the maximum amplitude ASc, and the waveform in the period Tg is This corresponds to the peak Sg that maximizes the amplitude ASg.

  The pulse generation unit 17 extracts a period in which an amplitude exceeding a predetermined threshold value continues from the myoelectric signal smoothed from the low-pass filter 16 as one pulse signal. An example of the pulse generator 17 is shown in FIG. The pulse generation unit 17 includes a threshold value memory 171 that holds a threshold value of the amplitude of the myoelectric potential signal, and a comparator 172 that compares the threshold value held in the threshold value memory 171 with the amplitude value of the detected myoelectric potential signal. Prepare. The pulse generation unit 17 includes a pulse output circuit 173 that outputs a pulse signal according to the comparison result in the comparator 172.

  The threshold value may be a fixed setting value, or the threshold value setting may be changed. In this case, as shown in FIG. 5, the pulse generation unit 17 includes a control unit 174 as threshold setting means for changing the threshold held in the threshold memory 171, and the comparator 172 sets the amplitude set by the control unit 174. This threshold value may be compared with the detected myoelectric potential signal. The control unit 174 executes control for changing the threshold value according to an input made by the user via an operation unit (not shown).

  Further, the signal processing device 1 may perform trial detection of myoelectric potential and determine a threshold value used for the subsequent main detection. Various methods can be used for setting the threshold, but any method may be used. Depending on the characteristics of the low-pass filter 16 used for smoothing the myoelectric signal, it is preferable to optimize the threshold value so that the pulse width is appropriately detected.

  FIG. 6 shows a pulse signal generated from the myoelectric potential signal with reference to the threshold value. 6A shows the myoelectric potential signal smoothed by the low-pass filter 16, and FIG. 6B shows the pulse signal generated by the pulse generator 17. In FIG. 6, the pulse generator 17 converts a peak Sa having a maximum exceeding the amplitude threshold Ath into a pulse signal Pa having a period Ta, converts the peak Sc into a pulse signal Pc having a period Tc, and the peak Se It is converted into a pulse signal Pe having a period Te, and the peak Sf is converted into a pulse signal Pf having a period Tf.

  In order to convert a myoelectric potential signal into a pulse signal, in addition to digital processing as described above, a part or all of the signal may be realized by an analog circuit using an analog filter, an integration circuit, an analog comparator, or the like.

  In the signal processing apparatus 1 having the above-described configuration, the myoelectric potential signal detected by the myoelectric potential detection sensor 11 is amplified by the biological amplifier 12 and converted from an analog signal to a digital signal by the A / D conversion circuit 13. The converted myoelectric potential signal is subjected to the removal of DC component noise by the high-pass filter 14, then the absolute value is taken, and further smoothed by the low-pass filter 16. The smoothed myoelectric potential signal is converted into a binary pulse signal by the pulse generator 17 and output from the output terminal 18.

  Since the signal processing device 1 does not extract the waveform of the myoelectric signal itself, which is one of the bioelectric signals, but converts the myoelectric signal into a pulse signal, the myoelectric signal is handled as a pulse signal. There is an advantage that the reproducibility of the myoelectric signal output as a pulse signal does not depend on the recognition rate of the pulse signal.

  Next, an example in which the signal processing device 1 described above is applied to a configuration for inputting a control signal of a mechanical device will be described with reference to FIG. Here, an example applied to a toy connected to an example applied to a radio control car (hereinafter referred to as a radio controlled car) in which a controller and a main body are remotely operated by wireless communication as an example of a mechanical device will be described.

  The radio controlled car 100 includes a controller 20 and a main body 30. The controller 20 includes a myoelectric potential detection sensor 21 as a bioelectric signal acquisition means for detecting a myoelectric potential signal that is one of bioelectric signals resulting from voluntary movement, a pulse detection unit 22, and a detected pulse pattern. A pulse pattern detection unit 23 to identify, a control signal generation unit 24 that generates a control signal corresponding to a control command from the identified pulse pattern, a modulation and transmission unit 25 that modulates the control signal and transmits it to the main body unit 30; And an antenna 26. These configurations are comprehensively controlled by the controller control unit 27. The controller 20 is a type having two myoelectric potential detection sensors, and can independently obtain pulse signals from myoelectric potential signals detected by one myoelectric potential detection sensor. And the control command of the main-body part 30 is assigned with respect to the combination of the pulses detected by two sensors. In the figure, the configurations corresponding to the two sensors are distinguished by adding a and b after the number.

  The main body 30 receives the radio wave transmitted from the controller 20 via the antenna 31, receives and demodulates the control signal from the received radio wave, generates a control command from the control signal, and drives the drive unit 34. A control signal generation unit 33 that generates a drive control signal for control, and a drive unit 34 that includes a motor that serves as power, a servo that changes the direction of the wheels, and the like. These components are comprehensively controlled by the main body control unit 35.

  In the controller 20, the myoelectric potential detection sensor 21 and the pulse detection unit 22 correspond to the signal processing device 1 described above. Accordingly, the pulse detection unit 22 includes a biological amplifier 12 that amplifies the detected myoelectric potential signal, an A / D conversion circuit 13 that converts the amplified analog myoelectric potential signal into a digital signal, and a high-pass for noise cut. A filter 14, a full-wave rectification circuit 15 for rectifying the myoelectric potential signal, a low-pass filter 16 for smoothing the myoelectric potential signal detected by the myoelectric potential detection sensor 11, and a threshold value among the smoothed myoelectric signals. A pulse generation unit 17 that extracts a period in which myoelectric potential signals having an amplitude exceeding the amplitude are extracted as one pulse signal, and an output terminal 18 that outputs a binarized pulse signal are provided.

Pulse pattern detector 23, by the difference between the rise time and the fall time of the pulse signal generated in the pulse detection unit 22 detects the pulse width, detects a pulse interval between adjacent pulse signals, The time series pattern of the pulse signal from the pulse detector 22 is detected.

  An example of a pulse pattern detection method is shown in FIGS. As shown in FIG. 8A, the pulse pattern detection unit 23 sets a pattern A when two pulses P1 and P2 are detected within a predetermined time Td, and generates one pulse P1 as shown in FIG. 8B. Is detected, it is classified as pattern B. Further, as shown in FIG. 9A, the pattern C is classified when the pulse width is within Ts, and the pattern D is classified when the pulse width is equal to or greater than Tl. For example, when the rising time of the first pulse signal P1 is T1, the falling time T2, the rising time of the second pulse signal P2 is T3, and the falling time is T4, (T3-T1) , (T3-T2), (T4-T1), and (T4-T2).

  The pulse pattern detection unit 23 can increase the types of patterns that can be detected by appropriately combining both the pulse width and the pulse interval. In this specific example, since such pulse pattern detection is simultaneously performed on signals obtained by the two myoelectric potential detection sensors, the types of control commands to be assigned can be increased. At this time, in order to prevent detection contrary to the user's intention due to subtle deviations in timing, processing that is regarded as simultaneous if a pulse signal appears within a certain time width range, even if it is not completely simultaneous Is more practical.

  Specifically, in this example, the pulse pattern detection unit 23 detects the fixed period Td as 0.7 seconds and a pulse width within 0.3 seconds as the pulse width of one pulse signal. For example, as shown in FIG. 10A, when a pulse having a time width of less than 0.3 seconds is detected twice within 0.7 seconds, pattern A is obtained, and as shown in FIG. When a pulse having a time width of less than 0.3 seconds is detected only once within 0.7 seconds, pattern B is used. As shown in FIG. 10C, a pulse having a time width of 0.7 seconds or more is detected. If detected, pattern C is assumed. A total of three types of pulse patterns are prepared, and this combination is made to correspond to the control command.

  FIG. 11 shows an example of a control command for the operation of the radio controlled car assigned to the pulse pattern. A combination of pulses detected from the two myoelectric potential detection sensors 21a and 21b is associated with a command for driving the main body. When a pulse signal is detected from the two myoelectric potential detection sensors 21a and 21b, a stop command is first issued. When the pulse pattern generated from the myoelectric potential detected by the myoelectric potential detection sensor 21a is the pattern A, the pulse pattern is generated from the myoelectric potential detected by the myoelectric potential detection sensor 21b. When the pattern is pattern A, the leftward movement is performed. Further, when the pulse pattern generated from the myoelectric potential detected by the myoelectric potential detection sensor 21a and the pulse pattern generated from the myoelectric potential detected by the myoelectric potential detection sensor 21b are both pattern A, the pattern is retreated. At that time. Further, when the pulse pattern generated from the myoelectric potential detected by the myoelectric potential detection sensor 21a and the pulse pattern generated from the myoelectric potential detected by the myoelectric potential detection sensor 21b are both patterns C, a command to sound a horn is used. . The combination of pulse patterns represented by hyphens in FIG. 11 is not used in this example, but another control command may be assigned.

  The control signal generation unit 24 generates a control signal corresponding to the control command from the pulse pattern as described above, and sends the control signal to the modulation and transmission unit 25. The modulation and transmission unit 25 modulates the control signal and transmits it to the main body unit 30 via the antenna 26. The main body 30 receives the radio wave transmitted from the controller 20 via the antenna 31, and the reception and demodulation unit 32 demodulates the control signal from the received radio wave. The control signal generation unit 33 generates a control command from the control signal, generates a drive control signal for controlling the drive unit 34, and sends the drive control signal to the drive unit 34. The drive part 34 changes the direction of a motor or a wheel based on the sent control signal.

  Although the mounting positions of the electrodes of the two myoelectric potential detection sensors of the controller 20 are arbitrary, for example, by attaching the myoelectric potential detection sensor 21a to the right arm and the myoelectric potential detection sensor 21b to the left arm, the radio controlled car by the movement of the arm 100 control approaches the handle operation interval and becomes more intuitive. Further, if sensor electrodes are attached to the shallow finger flexors of the forearm, the myoelectric potential associated with the bending of the fingers can be detected, so that the main body 30 can be operated simply by bending the left and right fingers.

  As described above, the signal processing device 1 does not extract the waveform of the myoelectric potential signal itself, which is one of the bioelectric signals, as described above, but converts the myoelectric potential signal into a pulse signal and uses it as a pulse signal. By handling, there is an advantage that the reproducibility of the myoelectric potential signal does not depend on the recognition rate of the pulse signal. Therefore, the radio controlled car 100 to which the signal processing device 1 is applied enables an input operation that is more intuitive and easy to understand for the user.

  The present invention can be applied as an input device of an electronic device. By generating a binary pulse signal from a bioelectric signal waveform caused by voluntary movement, a new entertainment application can be provided to the user and a new technical application can be created.

It is a block diagram explaining the signal processing apparatus shown as a specific example of this invention. It is a figure showing the waveform of the myoelectric potential signal A / D converted by the A / D conversion circuit in the said signal processing apparatus. It is a figure showing the waveform of the myoelectric potential signal smooth | blunted with the low-pass filter in the said signal processing apparatus. It is a block diagram explaining an example of the pulse generation part of the said signal processing apparatus. It is a block diagram explaining the other example of the pulse generation part of the said signal processing apparatus. (A) is a figure showing the myoelectric potential signal smooth | blunted with the low-pass filter, (b) is a figure showing the pulse signal produced | generated in the pulse generation part. It is a figure explaining the example which applied the said signal processing apparatus to the radio control car. It is a figure explaining the pulse pattern produced | generated by the said signal processing apparatus, (a) is a figure which shows the case where there are two pulses in a predetermined period, (b) is one pulse in a predetermined period It is a figure which shows a case. It is a figure explaining the pulse pattern produced | generated by the said signal processing apparatus, (a) is a figure which shows the case where the pulse which has the pulse width in period Ts is made into 1 pulse, (b) is a figure more than period Tl It is a figure which shows the case where the pulse which has a pulse width is made into 1 pulse. (A) is a figure explaining the case where one control command is assigned to two pulses generated within a predetermined period, and (b) is the case where one control command is assigned to one pulse generated within a predetermined period. (C) is a figure explaining the case where one control command is allocated with respect to the pulse which has a pulse width more than a predetermined period. It is a figure explaining the example of the control command which controls operation | movement of the toy allocated with respect to the pulse pattern.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Signal processing apparatus, 11 Myoelectric potential detection sensor, 12 Biological amplifier,
13 A / D conversion circuit, 14 High-pass filter, 15 Full-wave rectification circuit,
16 low-pass filter, 17 pulse generator, 18 output terminal,
20 controller, 21 myoelectric potential detection sensor, 22 pulse detection unit,
23 pulse pattern detection unit, 24 control command generation unit,
25 modulation and transmission unit, 26 antenna, 27 controller control unit,
30 body part, 31 antenna, 32 reception and demodulation part,
33 control signal generation unit, 34 drive unit, 35 main body control unit,
171 threshold memory, 172 comparator, 173 pulse output circuit,
174 Control unit

Claims (10)

Bioelectric signal acquisition means for acquiring a bioelectric signal resulting from voluntary movement;
Smoothing means for smoothing the bioelectric signal acquired by the bioelectric signal acquiring means;
Threshold holding means for holding a threshold of the amplitude of the bioelectric signal;
Comparison means for comparing the amplitude threshold held in the threshold holding means and the amplitude of the bioelectric signal acquired in the bioelectric signal acquisition means;
Pulse signal generating means for extracting, as one pulse signal, a period of continuous bioelectric signals having an amplitude exceeding the threshold value from the comparison result in the comparing means among the bioelectric signals smoothed in the smoothing means;
Pulse signal output means for outputting the pulse signal generated in the pulse signal generation means ;
A pulse for detecting the difference between the rise time and the fall time of the pulse signal generated by the pulse signal generation means as a pulse width and detecting the pulse interval between adjacent pulses of the pulse signal generated by the pulse signal generation means signal processing device Ru and a pattern detection means. The pulse pattern detection means detects whether the pulse width of the pulse signal generated by the pulse signal generation means is within a certain time Ts or more than a certain time Tl, and a pulse with a pulse width within the certain time Ts is detected within a certain time Td. The signal processing apparatus according to claim 1, wherein the type of pattern is discriminated based on a combination with the determined number . The bioelectric signal is a myoelectric signal,
The pulse pattern detection means uses a pattern A when a pulse having a pulse width of 0.3 seconds or less generated by the pulse signal generation means is detected twice within a period of 0.7 seconds. Pattern B is when a pulse with a width within 0.3 seconds is detected once within a period of 0.7 seconds, and pattern C is when a pulse with a pulse width of 0.7 seconds or more is detected the signal processing apparatus according to claim 1 or 2, wherein to determine the type. The pulse pattern detection means, when the rising time of the first pulse signal T1, the fall time T2, also the rising time of the second pulse signal T3, the falling time and T4, (T3-T1), (T3-T2), (T4 -T1), (T4-T2) signal processing apparatus according to any one of claims 1 to 3 that detect the pulse interval between adjacent pulse signals based on either . Comprising a plurality of said bioelectrical signal acquisition means, pulse signal generating means, claim that to extract period bioelectrical signal having an amplitude that exceeds the threshold value for each of a plurality of bioelectric signals consecutively in one pulse signal 5. The signal processing device according to any one of 1 to 4 . Bioelectric signals due to the voluntary movements, the signal processing apparatus according to claim 1 or 2, wherein Ru oculogram signals der. A bioelectric signal acquisition step of acquiring a bioelectric signal resulting from voluntary movement;
A smoothing step of smoothing the bioelectric signal acquired in the bioelectric signal acquisition step;
A comparison step of comparing the amplitude threshold held in the threshold holding means holding the amplitude threshold of the bioelectric signal with the amplitude of the bioelectric signal acquired in the bioelectric signal acquisition step;
A pulse signal generation step of extracting, as a one-pulse signal, a period of continuous bioelectric signals having an amplitude exceeding the threshold value from the comparison result in the comparison step among the bioelectric signals smoothed in the smoothing step;
A pulse signal output step of outputting the pulse signal generated in the pulse signal generation step ;
The difference between the rise time and the fall time of the pulse signal generated in the pulse signal generation process is detected as a pulse width, and the pulse interval between adjacent pulses of the pulse signal generated in the pulse signal generation process is detected. signal processing method that having a pulse pattern detection step of. In the pulse pattern detection step, whether the pulse width of the pulse signal generated in the pulse signal generation step is within a certain time Ts or more than a certain time Tl, and a pulse with a pulse width within the certain time Ts is within a certain time Td. The signal processing method according to claim 7, wherein the type of pattern is determined based on a combination with the detected number . The bioelectric signal is a myoelectric signal,
In the pulse pattern detection step, a pattern A is defined when a pulse with a pulse width of 0.3 seconds or less generated in the pulse signal generation step is detected twice within a period of 0.7 seconds, When a pulse with a pulse width of 0.3 seconds or less is detected once within a period of 0.7 seconds, it is referred to as pattern B, and when a pulse with a pulse width of 0.7 seconds or more is detected as pattern C, The signal processing method according to claim 7 or 8 , wherein the type of pattern is determined . In a mechanical device whose operation is controlled based on a control signal input from a user,
Bioelectric signal acquisition means for acquiring a bioelectric signal resulting from voluntary movement;
Smoothing means for smoothing the bioelectric signal acquired by the bioelectric signal acquiring means;
Threshold holding means for holding a threshold of the amplitude of the bioelectric signal;
Comparison means for comparing the amplitude threshold held in the threshold holding means and the amplitude of the bioelectric signal acquired in the bioelectric signal acquisition means;
Pulse signal generating means for extracting, as one pulse signal, a period of continuous bioelectric signals having an amplitude exceeding the threshold value from the comparison result in the comparing means among the bioelectric signals smoothed in the smoothing means;
Storage means for storing the signal pattern of the pulse signal and the control signal in association with each other;
Pulse signal output means for outputting the pulse signal generated in the pulse signal generation means ;
A pulse for detecting the difference between the rise time and the fall time of the pulse signal generated by the pulse signal generation means as a pulse width and detecting the pulse interval between adjacent pulses of the pulse signal generated by the pulse signal generation means Pattern detection means;
Mechanical Ru and control means for performing control to change the operation according to the control signal a control signal selected from the memory means corresponding to the pulse signal pattern output from the pulse pattern detecting means device.

Images