Search

The instantaneous frequency of a sine wave can be estimated using a Hilbert transformer. However, there is a ripple in the amplitude characteristics of finite-order Hilbert transformers. The estimate thus contains an oscillatory component that depends on this ripple. In this paper, we theoretically show that the frequency of the oscillation component is an even multiple of the input signal frequency. We propose a method for designing a low-pass FIR filter with variable transmission zeros. A simulation is used to show that this filter improves estimation accuracy.

In the visual inspection, the quality assurance is difficult, because the dispersion occurs in the result by skill and fatigue degree of the inspector. Recently, a visual inspection method by image processing using deep learning has been proposed. When using deep learning, the dataset to be used is important. In this paper, we describe a method for detecting painting defects using image processing, automatically generating data for deep learning, and using these data for classification using deep learning.

Frequency estimation is used in a wide range of areas such as radar, telecommunications, voice analysis and weight measurement using tuning fork sensor. This paper describes an instantaneous frequency (IF) estimation method. The original signal combined with the Hilbert transform form a complex analytic signal. The phase of the analytic signal is defined as the instantaneous phase and the time derivative of its phase IF. However, the amplitude of the signal obtained by phase-shifting the input signal by 90 degrees is different from that of the input signal because a ripple is generated in the frequency characteristics of the finite-order finite impulse response (FIR) Hilbert transformer (HT). As a result, errors are contained in the IF. In this paper, we propose a highly accurate IF estimation method using a notch filter. It is shown analytically that a harmonic frequency component is included in the IF obtained by using a finite-order FIR HT. Then, it is shown that the accurate instantaneous frequency can be obtained by removing the harmonic frequency component using an infinite impulse response (IIR) notch filter. Finally, the effectiveness of the proposed method is shown in simulations.

To measure mass of goods, automatic checkweighers are widely used, and is composed by a sensor of scale and a conveyor-belt. Variable digital filters (VDFs) are used to reduce the noise overlapping on the measurement signals in the instruments. The VDFs have piecewise high attenuation in the stopband to reduce delay. Conventional VDFs were difficult to obtain the desired piecewise high attenuation in order to change the attenuation by weight function. In this paper, we are proposed a design method for equiripple linear-phase FIR filters having multiple variable components in the stopband and the specified piecewise high attenuation. The proposed method adds a constraint in the minimax design problem, that directly changes piecewise high attenuation band by a variable parameter. The optimization methodology known as semi-definite programming (SDP) is used. The obtained VDF has high attenuation in part of the stopband, and then its attenuation, position, and stopband edge frequency are variable. Effectiveness of the proposed method is illustrated by a design example.

Circulating tumor cells (CTCs) can be used for the early detection of cancer and determination of the effectiveness of anticancer drug treatment. Sometimes CTC grows on a culture medium which has a honeycomb texture, and existing detection methods cannot achieve satisfactory results because the honeycomb textures are more salient than the CTCs edges. Thus, a CTCs extraction method for honeycomb textures from microscope images is needed.In this paper, we propose a new algorithm to remove honeycomb texture from microscope images. We confirmed that our proposed algorithm can sufficiently remove the honeycomb texture even the periodicity of the patterns are weak. Also our method achieves a high segmentation accuracy for images where the background is more salient than the CTCs edges.

Drowsiness can reduce working efficiency and result in dozing while driving. It is known that short naps in sleep stage 2 is effective for eliminating drowsiness, so further analysis of sleep stage 2 are required to clarify the relationship between nap length and quality. EEG in sleep stage 2 is characterized by the appearance of a sleep spindle and detection of this sleep spindle is thus required for analysis of this sleep stage. In this paper, we propose a method to detect a sleep spindle in the time domain using a Long Short-Term Memory (LSTM) network.

In general, a design of digital filters is used computers having relatively long bit length. On the other hand, filter coefficients could be rounded to the specified number of bit length, when digital filters are implemented embedded equipment. Then, filter characteristics may become degradation. It is difficult to realize high accuracy measurement. In this paper, we propose a finite bit length design method of variable digital filters with variable multiple elements of stopband. Proposed variable filter's coefficients are approximated by polynomial using variable parameters, and the polynomial coefficients are composed by finite bit length. The design problem is multiple objective optimization due to variable multiple elements. Then, we propose individual evaluation method for design variable digital filters using a genetic algorithm. Finally, effectiveness of the proposed method is shown by a design example.

It is known that sound signals dominated by howling noise have a large amplitude and a stable phase delay at a specific frequency. However, conventional howling cancellers focus only on the amplitude information. This paper proposes a howling canceller based on phase delay information. First, the group delay of the signal including howling is analyzed by short-time Fourier transform, and is determined that the variance of the group delay at the howling frequency is smaller than that at the speech signal frequency. Next, a new howling cancellation system is designed by introducing the variation in the group delay into the formula used to update adaptive notch filter. Finally, it is demonstrated via simulation that the proposed howling canceller can more effectively remove howling than can conventional howling cancellation methods.

In this paper, we propose a method for designing a low-delay low-pass/band-pass maximally flat integer-order differentiators in passband based on flatness characteristics and in stopband based on equiripple characteristics. The proposed function consist of difference between two functions. One of functions has low-delay and flat characteristics in the passband. The other gives the whole function equiripple characteristics in the stopband. The low-delay low-pass/band-pass integer-order differentiators designed by the proposed method realizes highly accurate differential estimation near the desired frequency and sharp cutoff characteristics.

Recently, Wang et al. proposed a general structure for maximally flat FIR digital filters, which can dynamically adjust frequency response only by changing derivative constraints. However, Wang's structure requires large circuit scale to realize the variable structure. To reduce the circuit scale, the present paper proposes a novel general structure for such filters that can also dynamically adjust frequency response. The proposed structure can realize exactly the same frequency response as Wang's structure with fewer multipliers, adders, and delays. Furthermore, the proposed structure can realize more types of frequency response than Wang's structure.

Afternoon sleepiness in daily life reduces arousal level, performance, and so on. It has been cleared that short naps are effective to cancel the sleepiness. Sleep stage 2 is one of important factors about sleeping especially in short time nap. Sleep spindles are especially important hallmarks of sleep stage 2. Therefore, it is necessary to find a spindle for analysis in sleep stage 2. In this paper, we propose an enhanced sleep spindle detector in time domain. In the proposed method, the sleep spindle waves are detected using the improved Fujimori's method and cumulative sum chart. Through example, the proposed method can detect the sleep spindles with 97.32 % sensitivity compared with a medical specialist. The proposed method can also extract theta and delta waves.

Afternoon sleepiness is a routine physiological phenomenon and it reduces arousal level, performance, and so on. It has been cleared and is often said that taking a short nap is effective to cancel the sleepiness. Sleep stage 2 is one of important factors on sleeping especially in short naps. Also, sleep spindles are important hallmarks of EEG in sleep stage 2. Therefore, it is necessary to find a spindle for analysis in sleep stage 2. In this paper, we propose a sleep spindle detector in time domain. In the proposed method, the sleep spindle waves are detected using the Fujimori's method and cumulative sum chart with occurrence tendency of candidates. Through example, the proposed method can detect the sleep spindles with 91.22 % sensitivity compared with a medical specialist.

Classifying leukocyte and examining their proportions is a very important for disease examination and diagnosis. This examination needs the knowledge of experts and a lot of time. Therefore, many automatic leukocyte images classification algorithms have been proposed. There is a method to classify 13 types blood cells using One versus One Support Vector Machine in one of them. In the conventional classify 13 types blood cells using One versus One Support Vector Machine in one of them. In the conventional method, leukocyte images are classified with the 26-dimensional feature vectors. The feature vectors are composed of 18-dimensional feature vectors relating to the color of cell and 8-dimensional feature vectors relating to the form of cell. The classification accuracy, however, is poor with these feature vectors regarding immature blood cells in this method. In this paper, we propose new feature vectors to improve the classification accuracy of immature blood cells with low classification accuracy among the leukocyte fractions. That is, we add two feature vectors in the proposed method.

We treat the design problem of second-degree IIR digital differentiators. The design problem (cost function) is formulated in the quadratic form without any frequency sampling. Since the cost function is the quadratic form, the solution is unique and the optimization scheme does not require any recursive optimization. Hence, the procedure of the design for the second-degree differentiators is very easy. Also, the second-degree differentiators obtained by our method have a robust stability since the maximum pole radius can be prescribed. Finally, we show a design example in order to demonstrate the effectiveness of the proposed method.

In this paper, we propose a design method of the low delay maximally flat band-pass FIR digital differentiators (DDs) with a stopband based on the L p norm criterion. Conventional low-delay band-pass FIR DDs realize maximally flat characteristics over the full range of the signal. However, the proposed DDs realize stopband based on the L p norm criterion and low-delay simultaneously. Furthermore, the proposed DDs can also realize linear phase characteristics. In this paper, a design problem of the transfer function is substituted for a design problem of the inverse characteristics function. Therefore, it is easy to derive a function that satisfies the flatness conditions. The passband function of the inverse characteristics function is based on the L p norm criterion. However, we presented a straightforward method, in which minimized the L p norm of the complex plane using Newton's method.

Two-dimensional (2-D) linear phase maximally flat finite impulse response (FIR) digital filters have flat passbands and stopbands. Conventionally, there are only fan-shaped or diamond-shaped. In this paper, we propose a method for designing 2-D low-pass maximally flat FIR digital filters that can realize various passband shapes. The proposed transfer function is achieved as a closed-form expression, and various passband shapes can be created by changing the filter order or flatness degree.

Design methods for digital filters with sparse coefficients (0 coefficients) are beneficial for reduction of arithmetic cost. Although the number of multipliers of the filter (arithmetic cost) can be reduced by using 0 coefficients, the performance of the filter can be degraded. In this paper, we propose a design method for multi-band digital filters with not only the sparse coefficients but also the sameness conditions of filter coefficients in which the absolute values of some coefficients are same. It is expected that the performance of the filter with the sameness conditions of filter coefficients can be better than that with just the sparse coefficients. Since the design problem is to optimize the combination of the sameness conditions and compute the real values of filter coefficients, the design problem is more complicated than that of the sparse filters. In order to optimize the combination of the sameness conditions, we employ a genetic algorithm (GA). Also, the real coefficients can be computed based on the Lagrange multiplier method. Finally, we show a design example in order to demonstrate the effectiveness of the proposed method.

Vibrating sensors are used as a sensor to measure such as pressures, forces, accelerations and densities. In particular, a tuning fork sensor is suitable as a sensor for a electronic balance due to high sensitivities to the force and its long term stabilities. However, the tuning fork sensor is affected by a temperature and an air density. Therefore, the tuning fork sensor should be compensated the effects of them. In a conventional method, the tuning fork sensor could not be compensated adequately for an abrupt change in the air density due to a delay of a smoothing filter. In this paper, we propose accurate compensation method of a drift of the tuning fork sensor which is caused by a changing the air density. Applying Kalman filter, the drift of the tuning fork sensor can be very small even when the abrupt change in the air density and there is a noise in a measured air pressure and its frequency.

This paper describes a design of linear phase band-pass FIR digital differentiators with flat passband and Lp (1 ≤ p ≤ ∞) norm-based stopband characteristics. The transfer function consists of a passband function and a stopband function. The passband function guarantees the flat differential characteristics around an arbitrary center frequency. In addition, the maximally flat differentiators provide highly accurate differentiation around the desired frequency (the center frequency). The stopband function provides the Lp norm-based stopband while correcting the stopband region of the passband function. The coefficients of the stopband function are obtained through the quasi-minimization based on IRLS (Iterative Reweighted Least Squares) algorithm. The proposed design achieves the flexible stopband under the trade-off between the transition bandwidth and the size of side-lope ripple.

The low-pass maximally flat FIR digital differentiator (low-pass MFDD) provides high accurate differentiation around the frequency ω = 0, while it reduces noises in the high frequency. Recently, authors proposed the low delay low-pass MFDD, but it can not get the steep cutoff characteristic. This paper proposes a design method of the low delay low-pass FIR digital differentiators having the flat passband and the equiripple stopband. The proposed transfer function is consist of the passband function and the stopband function. The passband function realizes the high accurate differentiation and the desired delay at ω = 0. The stopband function provides arbitrary roots in the stopband. Those roots are updated to be the optimal ones by using Remez algorithm iteratively.

In this paper, we propose a new frequency estimation method for a sinusoidally modulated signal. The proposed method is constructed by three processing; the noise reduction, the frequency estimation, the updating of the noise reduction filter. In the noise reduction, we utilize an adaptive IIR bandpass filter which can extract only a specific frequency components. In the frequency estimation, we employ DESA-2, which is a non-linear operator with low computational complexity. The extraction frequency of the bandpass filter is effectively updated by integrating the estimation results of DESA-2 and the gradient estimation method. Through the estimation simulation for the noisy fluctuate sinusoidal signal, we show that the proposed method can estimate the instantaneous frequency accurately.

This paper describes a general expression of the transfer function of the low-pass maximally flat FIR digital differentiator (low-pass MFDD). The low-pass MFDD provides extremely high accurate differentiation around ω = 0. Simultaneously, it can reduce noises in the high frequency. However, the conventional design method of the low-pass MFDD is designed with the linear phase characteristics. By contrast, the proposed method can adjust the value of the group delay at ω = 0, so that the proposed method can realize not only the linear phase characteristics but the low or the high delay characteristics.

In this paper, we propose a design method for complex coefficients FIR variable digital filter (VDF) with changeable multicomponent in the stopband. The proposed VDF has piecewise high attenuation in the stopband, and the VDF varies some characteristics in the stopband using variable parameters. The complex filter coefficients are composed polynomial expression using variable parameters to vary the filter characteristics. The proposed method, the minimization problem for the complex coefficients FIR VDF with changeable multicomponent are formulated as semidefinite programming (SDP) in the frequency domain. Finally, we demonstrate the effectiveness of the proposed method.

This paper proposes a comb filter with adaptive notch bandwidth. The comb filter is used to reduce a periodic noise from an observed signal. To extract the desired signal completely, we should appropriately design the notch bandwidth of the comb filter. Specifically, we should design the notch bandwidth to be equal to the fluctuation bandwidth. In this paper, to automatically reduce only the periodic noise with the frequency fluctuation, we propose the comb filter which achieves the adaptive notch bandwidth. Simulation results show the effectiveness of the proposed comb filter.

In this paper, we propose a design method for variable IIR digital filters with changeable multifactors in the stopband using the successive projection (SP) method. The SP method uses less computer memory to design digital filters than semidefinite programming. The proposed variable IIR filter has piecewise high stopband attenuation in the stopband, and the filter varies stopband edge frequency, and both the frequency and its magnitude in the piecewise high attenuation. The filter coefficients are composed polynomial expression using variable parameters to vary the filter characteristics. Finally, we demonstrate effectiveness of the proposed method.

In this paper, we propose a new EEG autoanalysis system based on the Fujimori method in order to resolve the problems experienced when applying the Fujimori method to the EEG autoanalysis system. The Fujimori method involves the simulation of diagnosis, where EEG analysis specialists perform diagnosis using their eyes. However, the Fujimori method is not developed for use on a computer. Thus, there have been several reports on EEG autoanalysis using the Fujimori method. We have also developed a system based on the Fujimori method, but this system has several problems. First, the deterioration of frequency resolution at high frequencies. Second, the analysis is over a finite time, even though the system was developed for a clinical EEG test. Finally, underutilization of the advantage of the Fujimori method, which can be used for analysis in the time domain. Thus, the new system resolved the problems discussed in the present paper. First, by oversampling with the 4th band filter, the system improved the frequency resolution. Next, using ring buffers and binary files, the system can be used for real-time and infinite-time analysis. Finally, by creating a frequency graph next to the graph of the EEG and analysis results, we can determine the frequency of each wave at exactly the time at which the wave occurs.

In this paper, we propose a design method for two-channel nonuniform-division FIR filter bank with variable transmission zeros in the stopband at each filters. For changing transmission variable parameters. In addition, the proposed filter bank satisfies approximately perfect reconstruction condition, and exhibits quasi-equiripple characteristics in the passband and stopband. Linear programming is used to design the proposed filter bank in the frequency domain. The proposed variable two-channel nonuniform-division FIR filter bank can be implemented using the Farrow structure, which is suitable for real time signal processing. The usefulness of the proposed filter bank is demonstrated through an example.

Several application in digital signal processing demand variable digital filters with arbitrary variable frequency response. For example, in dynamic gravimetry, the main vibration noise is changed by kinds or size of measurement and spec of conveying means. In this case, if a part of main noise is changed, it is important that changing characteristics of filter at once. FIR digital filter are always stable. However, because FIR digital filter generally needs high order, the delay is increased and it is problem in high speed and high accuracy signal processing. Then this paper describes a design method for IIR digital filters with variable stopbands. Numerator filter coefficients are approximated by polynomials using a variable parameter. We solve its transfer function by semidefinite programming in the frequency domain. Finally, a design example is given to demonstrate the proposed techniques.

A applicability of Successive Projections (SP) method is wide. In the previous SP method which is an iterative approximation method, only one extreme frequency point at which the deviation from the given specification is maximized is used in the update of the filter coefficients. We extended to SP method which can update the filter coefficients by the multiple extreme frequency points using Fritz John's theorem. In this method, the solution is obtained less iteration number than the previous method. However, because it is necessary to decide the maximum allowable error experientially, there is the case that the solution does not converge and the obtained filter does not have equiripple characteristics. In this paper, we present a design method of linear phase FIR filters by SP method updating maximum allowable error. The updating maximum allowable error analytically solves such as Remez algorithm. Finally, we confirm the effectiveness of propose method through several examples.

In this paper, we propose a design method for finite impulse response (FIR) variable digital filters (VDFs) obtaining quasi-equiripple characteristic even if notch frequencies change. The change in filter coefficients, which depends on the change in notch frequencies, is approximated by a polynomial using variable parameters. In this method, the minimization problem for the polynomial coefficients with variable notches is formulated as semidefinite programming (SDP) in the frequency domain. However, the number of polynomial coefficients increases when variable parameters are plural number. Therefore, we also propose a method for reducing the number of polynomial coefficients. The usefulness of the proposed VDF is demonstrated through examples.

In this paper, we describe about the development of electric circuit exercise system that has interactiveness and effect of animation by using GUI operation, to make "Movement" of technological model easily underdtood for the students. Althogh we proposed iCass (Interactive Circuits and System Seminar) as a research similar to this paper, we advance the development of the educational support teaching material enhanced more by adding "Problem exercise" and "Learning assessment" insufficient in it.

In the field of measurement signal processing, we should often change the characteristic of the stopband instantaneously. Such the filter that can change the frequency response instantaneously is called variable filter. We propose the design of FIR digital filters with variable stopbands. In this paper, the quasi-chebyshev characteristic is obtained by using an iterative weighted least square (WLS) method. Moreover, a memory required for the calculation for computing filter coefficients in the proposed method has the feature of being fewer than the conventional method. The usefulness of the proposed design method is verified through some examples.

In this paper, we propose a transfer function of filters which does not change filter characteristics even if notch frequency changes. A change in filter coefficient which depends on change in a notch frequency is approximated by the polynomial which used a variable parameter. In this method, the minimization problems for the coefficients of the filter with a variable notch are formulated as semidefinite programming (SDP) in the frequency domain. Moreover we present that the proposed VDF can be implemented using the Farrow structure suitable for high-speed signal processing.

In this paper, we propose designing method for separable-denominator two-dimensional Infinite Impulse Response (IIR) filters (separable 2D IIR filters) by Successive Projection (SP) methods using a stability criterion based on the system matrix. It is generally known that separable 2D IIR filters are stable if and only if each of the denominators is stable. Therefore, the stability criteria of 1D IIR filters can be used for separable 2D IIR filters. A stability criterion based on the system matrix is a necessary and sufficient condition to guarantee stability in 1D IIR filters. Therefore, separable 2D IIR filters obtained by the proposed method have a smaller error ripple than those obtained by the conventional method.

In this paper, we define filters capable of changing frequency response characteristics as variable filters. Digital filters capable of changing their frequency response are such as group delay, magnitude response and resonance frequency, etc.. In this paper, we propose a design method for FIR digital filters with an arbitrary variable fractional delay by convex projections algorithm. Then each coefficients of variable filter represents as different polynomials of M order using a set of spectral parameters. Finally, the usefulness of the proposed method is verified through the examples.

Sampling rate conversion using the filter banks is proposed in the field of digital signal processing. However, in sampling rate conversion with a rational factor, the computational complexity may become very large. Moreover whenever the sampling rate is changed, it is necessary to design the filter again. In this paper, hence, we present a kernel with block structure for sampling rate conversion. As the filter proposed has the impulse response approximated by polynomials, it is unnecessary to redesign whenever the sampling rate is changed. The kernel proposed has the block structure that the impulse response of the sampling section since the third is represented by the polynomial used for the second sampling section. Therefore, the filter has less memory. Moreover, the filter has the advantage that it is possible to correspond to arbitrary fractional sampling rate conversion.

Several applications in digital signal processing demand variable digital filters with arbitrary variable frequency response. However, the design method of filters having variable piecewise large attenuation characteristics in the stopband is not proposed. This paper describes the design method of variable FIR digital filters having changed piecewise large attenuation characteristics in the stopband. The basic idea is to represent the coefficients of variable transfer function as the polynomials using spectral parameter. We design each filter with difference stopband attenuation by successive projections method. Then we approximate the obtained filter coefficient by the polynomials using spectral parameter in the least-square sense.

This paper presents a directly design method of FIR filters with minimum phase characteristics using successive projections (SP) method. We use Rouche's theorem for coefficients update algorithm in SP method to restrict the roots of transfer function inside the unit circle and on the unit circle. The proposed method is an iterative optimization technique and its algorithm is as simple as the conventional SP method. The usefulness of the proposed method is verified through the examples.