Random Noise Suppression Research Articles

Seismic Exploration Random Noise on Land: Modeling and Application to Noise Suppression

In seismic exploration, random noise suppression is one of the key problems in seismic data processing. For random noise attenuation, the most important thing is the understanding of seismic random noise generation and propagation. Seismic random noise is considered as temporal and spatial random processes, and it can be analyzed only qualitatively for now, due to its high variability. In this paper, we classify seismic random noise sources by their generation factors and simulate the random noise of the desert in West China. According to Green’s function, it can be assumed that seismic random noise sources are point-like sources that are distributed around geophones. A seismic random noise record is taken as the superimposed wave field exited by all the independent sources in a homogeneous isotropy half-infinite surface. Based on the wind vibration theory and preliminary study about ambient vibrations, the noise source functions are determined. We obtain the waveforms of different kinds of noise by solving the inhomogeneous wave equations and analyze the characteristics qualitatively and quantitatively. The seismic synthetic record with a 1.6-s time and 250-m distances is obtained, and the characteristics are compared between the simulated and the real noise record in time domain and space domain, respectively. The comparative results show the same characteristics of the simulated noise and the real noise, which demonstrates the feasibility of the proposed method. According to the noise modeling, it is known that the near-field cultural noise is the main component of the random noise in the desert, on the basis of which complex diffusion filtering is selected. The filtered results by complex diffusion filtering is compared with the results of time–frequency peak filtering, which is a popular filtering method of seismic random noise suppression in recent years. The comparative results show that complex diffusion filtering is more suitable for the noise of the desert in the Tarim Basin. This result proves that seismic random noise modeling can provide the guidance for noise attenuation. It lays a foundation for researching the propagation characteristics and better attenuation of seismic random noise in the future.

A robust detector for rolling element bearing condition monitoring based on the modulation signal bispectrum and its performance evaluation against the Kurtogram

Envelope analysis is a widely used method for rolling element bearing fault detection. To obtain high detection accuracy, it is critical to determine an optimal frequency narrowband for the envelope demodulation. However, many of the schemes which are used for the narrowband selection, such as the Kurtogram, can produce poor detection results because they are sensitive to random noise and aperiodic impulses which normally occur in practical applications. To achieve the purposes of denoising and frequency band optimisation, this paper proposes a novel modulation signal bispectrum (MSB) based robust detector for bearing fault detection. Because of its inherent noise suppression capability, the MSB allows effective suppression of both stationary random noise and discrete aperiodic noise. The high magnitude features that result from the use of the MSB also enhance the modulation effects of a bearing fault and can be used to provide optimal frequency bands for fault detection. The Kurtogram is generally accepted as a powerful means of selecting the most appropriate frequency band for envelope analysis, and as such it has been used as the benchmark comparator for performance evaluation in this paper. Both simulated and experimental data analysis results show that the proposed method produces more accurate and robust detection results than Kurtogram based approaches for common bearing faults under a range of representative scenarios.

Spectral Subtraction Denoising Preprocessing Block to Improve Slow Cortical Potential Based Brain–Computer Interface

The measured electroencephalographic (EEG) signals used in brain–computer interface (BCI) are usually contaminated with several additive random and physiological noise components. Even though several preprocessing strategies were proposed in the literature to extract useful EEG signal components for further processing and analysis, their performance is yet to meet practical needs to boost BCI applications performance. Of particular interest among those methods are the ones addressing random noise suppression that could help enhance the performance of low-cost BCI systems/headsets. A preprocessing block for signal denoising of slow cortical potential (SCP) that provides efficient noise removal and better classification accuracy is presented. This method is a variant of the spectral subtraction signal denoising whereby the noise power spectrum is estimated and removed adaptively. This method is applied to each EEG channel separately thus providing a preprocessing block that could be integrated with the existing spatial/temporal preprocessing methods. The impact of this method on classification accuracy is studied and compared to the conventional wavelet shrinkage method. The proposed method is verified using experimental data from the BCI competition II data set whereby a performance boost was obtained with the new preprocessing block. The improvement was quantitatively assessed using mean square error and mean absolute error measures and was also shown to be statistically significant. Moreover, spectral subtraction denoising is shown to have less computational complexity than wavelet shrinkage based methods. The new preprocessing block for SCP-based BCI signals based on spectral subtraction provides significant improvement in performance when added and offers an adaptive yet less computationally expensive alternative to existing methods such as wavelet shrinkage based denoising method. Also, given its independent channel processing, it shows potential for seamless integration into conventional processing chain for different BCI applications.

Application of the Radon–FCL approach to seismic random noise suppression and signal preservation

The fractal conservation law (FCL) is a linear partial differential equation that is modified by an anti-diffusive term of lower order. The analysis indicated that this algorithm could eliminate high frequencies and preserve or amplify low/medium-frequencies. Thus, this method is quite suitable for the simultaneous noise suppression and enhancement or preservation of seismic signals. However, the conventional FCL filters seismic data only along the time direction, thereby ignoring the spatial coherence between neighbouring traces, which leads to the loss of directional information. Therefore, we consider the development of the conventional FCL into the time-space domain and propose a Radon–FCL approach. We applied a Radon transform to implement the FCL method in this article; performing FCL filtering in the Radon domain achieves a higher level of noise attenuation. Using this method, seismic reflection events can be recovered with the sacrifice of fewer frequency components while effectively attenuating more random noise than conventional FCL filtering. Experiments using both synthetic and common shot point data demonstrate the advantages of the Radon–FCL approach versus the conventional FCL method with regard to both random noise attenuation and seismic signal preservation.

Noise suppressing and direct wave arrivals removal in GPR data based on Shearlet transform

We present a new application of Shearlet transform (ShT) for ground penetrating radar (GPR) to remove the direct wave arrivals and suppress random noise. In fact, the reflecting waves from the underground targets and the direct wave arrivals are sometimes overlapped when the targets are buried shallowly, which influences the arrival-time detection and the target-position location. Random noise is also one of the influencing factors. In this paper, ShT is introduced to remove the direct wave arrivals and suppress random noise. The original GPR data is transformed to ShT domain by using multi-scale and multi-direction properties of ShT. The direct wave arrivals and the remaining GPR signal are effectively separated. While we eliminate the direct wave arrivals, the GPR signal is not damaged. In ShT domain, GPR data can be expressed sparsely. The GPR signal is described by big Shearlet coefficients, but random noise is described by small ones. So we can use the threshold method depending on different scales and directions in ShT domain to suppress random noise. The GPR signal's signal to noise ratio (SNR) is enhanced. The synthetic and field GPR data verify the effectiveness of direct wave arrivals removal and random noise suppression based on ShT.

SUPPRESSION OF SEISMIC RANDOM NOISE BASED ON STEERABLE FILTERS

AbstractFor seismic random noise suppression, this work designs a steerable filter by taking advantage of elongated Hermite‐Gauss functions. According to the different directional responses between valid signal and random noise, we can reconstruct signal by the local characteristics of selected data. With the added directional selectivity, the filtering process can match different direction axes, which makes sure that noise is suppressed without reducing the signal fidelity. The property of directional steerability makes computation more efficient and requires less storage space. Simulation results show that we can get better signal amplitude and denoising effects than traditional wavelet transform and Curvelet transform algorithm by using this method. At –5 db SNR, this method can ensure that the average amplitude reaches 92.99% and SNR enhances 221.774%, which can significantly suppress noise as well as keep the useful signal in processing of real seismic signals.

SVD-Based Technique for Interference Cancellation and Noise Reduction in NMR Measurement of Time-Dependent Magnetic Fields.

A nuclear magnetic resonance (NMR) experiment for measurement of time-dependent magnetic fields was introduced. To improve the signal-to-interference-plus-noise ratio (SINR) of NMR data, a new method for interference cancellation and noise reduction (ICNR) based on singular value decomposition (SVD) was proposed. The singular values corresponding to the radio frequency interference (RFI) signal were identified in terms of the correlation between the FID data and the reference data, and then the RFI and noise were suppressed by setting the corresponding singular values to zero. The validity of the algorithm was verified by processing the measured NMR data. The results indicated that, this method has a significantly suppression of RFI and random noise, and can well preserve the FID signal. At present, the major limitation of the proposed SVD-based ICNR technique is that the threshold value for interference cancellation needs to be manually selected. Finally, the inversion waveform of the applied alternating magnetic field was given by fitting the processed experimental data.

Seismic directional random noise suppression by radial‐trace time–frequency peak filtering using the Hurst exponent statistic

ABSTRACTRadial‐trace time–frequency peak filtering filters a seismic record along the radial‐trace direction rather than the conventional channel direction. It takes the spatial correlation of the reflected events between adjacent channels into account. Thus, radial‐trace time–frequency peak filtering performs well in denoising and enhancing the continuity of reflected events. However, in the seismic record there is often random noise whose energy is concentrated in certain directions; the noise in these directions is correlative. We refer to this kind of random noise (that is distributed randomly in time but correlative in the space) as directional random noise. Under radial‐trace time–frequency peak filtering, the directional random noise will be treated as signal and enhanced when this noise has same direction as the signal. Therefore, we need to identify the directional random noise before the filtering. In this paper, we test the linearity of signal and directional random noise in time using the Hurst exponent. The time series of signals with high linearity lead to large Hurst exponent value; however, directional random noise is a random series in time without a fixed waveform and thus its linearity is low; therefore, we can differentiate the signal and directional random noise by the Hurst exponent values. The directional random noise can then be suppressed by using a long filtering window length during the radial‐trace time–frequency peak filtering. Synthetic and real data examples show that the proposed method can remove most directional random noise and can effectively recover the reflected events.

CHAMP climate data based on the inversion of monthly average bending angles

Abstract. Global Navigation Satellite System Radio Occultation (GNSS-RO) refractivity climatologies for the stratosphere can be obtained from the Abel inversion of monthly average bending-angle profiles. The averaging of large numbers of profiles suppresses random noise and this, in combination with simple exponential extrapolation above an altitude of 80 km, circumvents the need for a "statistical optimization" step in the processing. Using data from the US–Taiwanese COSMIC mission, which provides ~1500–2000 occultations per day, it has been shown that this average-profile inversion (API) technique provides a robust method for generating stratospheric refractivity climatologies. Prior to the launch of COSMIC in mid-2006, the data records rely on data from the CHAMP (CHAllenging Mini-satellite Payload) mission. In order to exploit the full range of available RO data, the usage of CHAMP data is also required. CHAMP only provided ~200 profiles per day, and the measurements were noisier than COSMIC. As a consequence, the main research question in this study was to see if the average bending-angle approach is also applicable to CHAMP data. Different methods for the suppression of random noise – statistical and through data quality prescreening – were tested. The API retrievals were compared with the more conventional approach of averaging individual refractivity profiles, produced with the implementation of statistical optimization used in the EUMETSAT (European Organisation for the Exploitation of Meteorological Satellites) Radio Occultation Meteorology Satellite Application Facility (ROM SAF) operational processing. In this study it is demonstrated that the API retrieval technique works well for CHAMP data, enabling the generation of long-term stratospheric RO climate data records from August 2001 and onward. The resulting CHAMP refractivity climatologies are found to be practically identical to the standard retrieval at the DMI (Danish Meteorological Institute) below altitudes of 35 km. Between 35 and 50 km, the differences between the two retrieval methods started to increase, showing largest differences at high latitudes and high altitudes. Furthermore, in the winter hemisphere high-latitude region, the biases relative to ECMWF (European Centre for Medium-range Weather Forecasts) were generally smaller for the new approach than for the standard retrieval.

Local spatiotemporal time–frequency peak filtering method for seismic random noise reduction

To achieve a higher level of seismic random noise suppression, the Radon transform has been adopted to implement spatiotemporal time–frequency peak filtering (TFPF) in our previous studies. Those studies involved performing TFPF in full-aperture Radon domain, including linear Radon and parabolic Radon. Although the superiority of this method to the conventional TFPF has been tested through processing on synthetic seismic models and field seismic data, there are still some limitations in the method. Both full-aperture linear Radon and parabolic Radon are applicable and effective for some relatively simple situations (e.g., curve reflection events with regular geometry) but inapplicable for complicated situations such as reflection events with irregular shapes, or interlaced events with quite different slope or curvature parameters. Therefore, a localized approach to the application of the Radon transform must be applied. It would serve the filter method better by adapting the transform to the local character of the data variations. In this article, we propose an idea that adopts the local Radon transform referred to as piecewise full-aperture Radon to realize spatiotemporal TFPF, called local spatiotemporal TFPF. Through experiments on synthetic seismic models and field seismic data, this study demonstrates the advantage of our method in seismic random noise reduction and reflection event recovery for relatively complicated situations of seismic data.

Application of Seismic Data Stacking in Time–Frequency Domain

Stacking is a fundamental concept that can be found in geosciences and seismic signal processing. The performance of random noise suppression and residual moveout correction by traditional amplitude-based stacking process is always far from satisfactory. In this letter, we present a time-frequency-domain phase-based stacking approach to exploit the similar spectral content of the different traces in seismic array data. The method can overcome the disadvantages of many existing stacking methods. The phase is derived from the well-known time-frequency-domain technique S transform, and the average of the phase spectrum is used to weight the spectrum of prestack seismic traces. As a coherency measure, the weighted phase spectrum can better reflect the lateral coherence of effective signal between individual traces. Meanwhile, the unique time-frequency properties of S spectrum make the weighting process more competitive in suppressing random noise and nonuniform artifacts. The analysis results of synthetic and real field data show the effectiveness of the proposed stacking approach.

Contourlet based seismic reflection data non-local noise suppression

Abstract In this paper, we propose a non-local, transform domain noise suppression framework to improve the quality of seismic reflection data. The original non-local means (NLM) algorithm measures similarities in the data domain and we generalize it in the nonsubsampled contourlet transform (NSCT) domain. NSCT gives a multiscale, multiresolution and anisotropy representation of the noisy input. The redundancy information in NSCT subbands can be utilized to enhance the structures in the original seismic data. Like the wavelet transform, NSCT coefficients in each subband follow the generalized Gaussian distribution and the parameters can be estimated using appropriate techniques. These parameters are used to construct our proposed NSCT domain filtering algorithm. Applications for synthetic and real seismic data of the proposed algorithm demonstrate its effectiveness on seismic data random noise suppression.

Suppression of strong random noise in seismic data by using time-frequency peak filtering

Time-frequency peak filtering (TFPF) is highly efficient in suppressing random noise in seismic data. Although the hypothesis of stationary Gaussian white noise cannot be fulfilled in practical seismic data, TFPF can effectively suppress white and colored random noise with different intensities, as can be theoretically demonstrated by detecting such noise in synthetic seismic data. However, a “zero-drift” effect is observed in the filtered signal and is independent of the average power and variance of the random noise, but related to its mean value. Furthermore, we consider the situation where the local linearization of the seismic data cannot be satisfied absolutely and study the “distortion” characteristics of the filtered signal using TFPF on a triangular wave. We found that over-compensation is possible in the frequency band for the triangular wave. In addition, it is nonsymmetrical and has a relationship to the time-varying curvature of the seismic wavelet. The results also present an improved approach for TFPF.

Seismic random noise suppression using an adaptive nonlocal means algorithm

Nonlocal means filtering is a noise attenuation method based on redundancies in image information. It is also a nonlocal denoising method that uses the self-similarity of an image, assuming that the valid structures of the image have a certain degree of repeatability that the random noise lacks. In this paper, we use nonlocal means filtering in seismic random noise suppression. To overcome the problems caused by expensive computational costs and improper filter parameters, this paper proposes a block-wise implementation of the nonlocal means method with adaptive filter parameter estimation. Tests with synthetic data and real 2D post-stack seismic data demonstrate that the proposed algorithm better preserves valid seismic information and has a higher accuracy when compared with traditional seismic denoising methods (e.g., f-x deconvolution), which is important for subsequent seismic processing and interpretation.

De-striping hyperspectral imagery using wavelet transform and adaptive frequency domain filtering

Hyperspectral imagers are built line-by-line similar to images acquired by pushbroom sensors. They can experience striping artifacts due to variations in detector response to incident imagery. In this research, a method for hyperspectral image de-striping based on wavelet analysis and adaptive Fourier zero-frequency amplitude normalization has been developed. The algorithm was tested against three other de-striping algorithms. Hyperspectral image bands of different scenes with significant striping and random noise, as well as an image with simulated noise, were used in the testing. The results were assessed visually and quantitatively using frequency domain Signal-to-Noise Ratio (SNR), Root Mean Square Error (RMSE) and/or Peak Signal-to-Ratio (PSNR). The results demonstrated the superiority of our proposed algorithm in de-striping hyperspectral images without introducing unwanted artifacts, yet preserving image details. In the noise-induced image results, the proposed method reduced RMSE error and improved PSNR by 3.5 dB which is better than other tested methods. A Combined method, integrating the proposed algorithm with a generic wavelet-based de-noising algorithm, showed significant random noise suppression in addition to stripe reduction with a PSNR value of 4.3 dB. These findings make the algorithm a candidate for practical implementation on remote sensing images including high resolution hyperspectral images contaminated with stripe and random noise.

An iterative curvelet thresholding algorithm for seismic random noise attenuation

In this paper, we explore the use of iterative curvelet thresholding for seismic random noise attenuation. A new method for combining the curvelet transform with iterative thresholding to suppress random noise is demonstrated and the issue is described as a linear inverse optimal problem using the L1 norm. Random noise suppression in seismic data is transformed into an L1 norm optimization problem based on the curvelet sparsity transform. Compared to the conventional methods such as median filter algorithm, FX deconvolution, and wavelet thresholding, the results of synthetic and field data processing show that the iterative curvelet thresholding proposed in this paper can sufficiently improve signal to noise radio (SNR) and give higher signal fidelity at the same time. Furthermore, to make better use of the curvelet transform such as multiple scales and multiple directions, we control the curvelet direction of the result after iterative curvelet thresholding to further improve the SNR.

Status and first results of the acoustic detection test system AMADEUS

The AMADEUS system is integrated in the ANTARES neutrino telescope in the Mediterranean Sea and aims for the investigation of acoustic particle detection techniques in the deep sea. Installed at a depth of more than 2000 m, the acoustic sensors of AMADEUS are using piezo-ceramic elements for the broad-band recording of acoustic signals with frequencies ranging up to 125 kHz. AMADEUS consists of six clusters, each one comprising six acoustic sensors that are arranged at distances of roughly 1 m from each other. Three acoustic clusters are installed along a vertical mechanical structure (a so-called Line) of ANTARES with spacings of about 15 and 110 m, respectively. The remaining three clusters are installed with vertical spacings of 15 m on a further Line of the ANTARES detector. The horizontal distance between the two lines is 240 m. Each acoustic cluster allows for the suppression of random noise by requiring local coincidences and the reconstruction of the arrival direction of acoustic waves. Source positions can then be reconstructed using the precise time correlations between the clusters provided by the ANTARES clock system. AMADEUS thus allows for extensive acoustic background studies including signal correlations on several length scales as well as source localisation. The system is therefore excellently suited for feasibility studies for a potential future large-scale acoustic neutrino telescope in sea water. Since the start of data taking on December 5th, 2007, a wealth of data has been recorded. The AMADEUS system will be described and some first results will be presented.

Fluorescence-Suppressed Raman Technique for Quantitative Analysis of Protein Solution Using a Micro-Raman Probe, the Shifted Excitation Method, and Partial Least Squares Regression Analysis

A practical Raman analyzing technique with suppression of the strong fluorescent background in order to obtain quantitative information is proposed in the present study. The technique is based on the shifted excitation method and partial least squares regression (PLSR) analysis. The Raman system consists of a single Raman spectrometer, a background-free electrically tunable Ti:Sapphire laser (BF-ETL), and a micro-Raman probe (MRP). The system allows one to obtain reliable shifted excitation Raman spectra with a simple operation. The PLSR analysis successfully provides quantitative information from the obtained spectra with the suppression of random noise including photon shot noise. The present study demonstrates that the technique is effective for extracting quantitative information concealed behind a fluorescent background that is more than 200 times stronger than the Raman signal.

Application of complex-trace analysis to seismic data for random-noise suppression and temporal resolution improvement

Time-dependent amplitude and phase information of stacked seismic data are processed independently using complex trace analysis in order to facilitate interpretation by improving resolution and decreasing random noise. We represent seismic traces using their envelopes and instantaneous phases obtained by the Hilbert transform. The proposed method reduces the amplitudes of the low-frequency components of the envelope, while preserving the phase information. Several tests are performed in order to investigate the behavior of the present method for resolution improvement and noise suppression. Applications on both 1D and 2D synthetic data show that the method is capable of reducing the amplitudes and temporal widths of the side lobes of the input wavelets, and hence, the spectral bandwidth of the input seismic data is enhanced, resulting in an improvement in the signal-to-noise ratio. The bright-spot anomalies observed on the stacked sections become clearer because the output seismic traces have a simplified appearance allowing an easier data interpretation. We recommend applying this simple signal processing for signal enhancement prior to interpretation, especially for single channel and low-fold seismic data.

Saving Bits—The Impact of MCTF Enhanced Noise Reduction

Motion Compensated Temporal Filtering (MCTF) has made a significant impact on the DTV industry by saving bits and allowing operators to deliver consistently higher quality video at lower rates than ever before. All content tends to contain some noise, the random and unwanted portion of a signal. Even modern facilities equipped with the latest digital production equipment will inevitably import noisy content unless the incoming material is thoroughly cleaned. Noise reduction (NR) and filtering can substantially improve the video received by a viewer if the right techniques are applied to remove noise prior to compression. Selectively removing noise is a challenge because it shares space with valuable picture detail. An ideal noise reduction process will allow powerful suppression of random noise while preserving clean video content. The major advantage of MCTF is its inherent ability to remove noise without introducing motion blur artifacts. This technique has been known for many years, but the costs associated with a high processing overhead has kept it from being commercially viable. This paper will explain how integration of the new MCTF technique with existing technologies has dramatically improved video-compression efficiency.