Use Of Matched Filter Research Articles

Matched Filter for Acoustic Emission Monitoring in Noisy Environments: Application to Wire Break Detection

Regular inspections of important civil infrastructures are mandatory to ensure structural safety and reliability. Until today, these inspections are primarily conducted manually, which has several deficiencies. In context of prestressed concrete structures, steel tendons can be susceptible to stress corrosion cracking, which may result in breakage of individual wires that is visually not observable. Recent research therefore suggests Acoustic Emission Monitoring for wire break detection in prestressed concrete structures. However, in noisy environments, such as wind turbines, conventional acoustic emission detection based on user-defined amplitude thresholds may not be suitable. Thus, we propose the use of matched filters for acoustic emission detection in noisy environments and apply the proposed method to the task of wire break detection in post-tensioned wind turbine towers. Based on manually conducted wire breaks and rebound hammer tests on a large-scale test frame, we employ a brute-force search for the most suitable query signal of a wire break event and a rebound hammer impact, respectively. Then, we evaluate the signal detection performance on more than 500 other wire break signals and approximately one week of continuous acoustic emission recordings in an operating wind turbine. For a signal-to-noise ratio of 0 dB, the matched filter approach shows an improvement in AUC by up to 0.78 for both, the wire break and the rebound hammer query signal, compared to state-of-the-art amplitude-based detection. Even for the unscaled wire break measurements originally recorded at the 12 m large laboratory test frame, the improvement in AUC still lies between 0.01 and 0.25 depending on the wind turbine noise recordings considered for evaluation. Matched filters may therefore be a promising alternative to amplitude-based detection algorithms and deserve particular consideration with regard to Acoustic Emission Monitoring, especially in noisy environments or when sparse senor networks are required.

Optimization of matched filters for VHR image segmentation

ABSTRACT The availability of very high resolution remotely sensed images has made the use of spectral information alone insufficient for class recognition, so the integration of spatial information into the segmentation task becomes necessary. Spatial information takes into account the surrounding of a pixel, rather than dealing with a pixel as an isolated item. In this paper, inspired by matched filtering theory, we propose a new technique for spatial feature extraction. The technique consists of using 2D kernels convolved with the spectral bands. The convolution operation bears valuable spatial information about the pixel under consideration. The concatenation of the extracted spatial features and spectral features of pixels feeds a support vector machine (SVM) to segment the image of interest. To cope with the complexity of images having objects of varying sizes (i.e. urban areas), we adopted a hierarchical strategy. That is, kernels with increasing size are applied to extract different spatial features corresponding to different objects. Thus, we have extended the use of matched filters from single object enhancement to multi-object enhancement. A challenging step in designing the matched filters is the two-dimensional kernels coefficients selection, which we formulated as an optimization problem within a particle swarm optimization framework. The optimization process was driven by two different fitness functions, the cross-validation SVM accuracy and the Bhattacharyya distance, which were both evaluated on training samples. We assessed the proposed procedure on two very high-resolution images having different spatial resolutions. The obtained results showed significant improvement in terms of the overall classification accuracy (over 10% for both images). Moreover, visual inspection of the segmented images, in addition to pepper and salt elimination, revealed significant improvement in many objects not detected by the spectral method. Our method of extracting spatial features seems to be very efficient.

P300 brainwave extraction from EEG signals: An unsupervised approach

The P300 is an endogenous event-related potential (ERP) that is naturally elicited by rare and significant stimuli, arisen from the frontal, temporal and occipital lobe of the brain, although is usually measured in the parietal lobe. P300 signals are increasingly used in brain-computer interfaces (BCI) because the users of ERP-based BCIs need no special training. In order to detect the P300 signal, most studies in the field have been focused on a supervised approach, dealing with over-fitting filters and the need for later validation. In this paper we start bridging this gap by modeling an unsupervised classifier of the P300 presence based on a weighted score. This is carried out through the use of matched filters that weight events that are likely to represent the P300 wave. The optimal weights are determined through a study of the data’s features. The combination of different artifact cancelation methods and the P300 extraction techniques provides a marked, statistically significant, improvement in accuracy at the level of the top-performing algorithms for a supervised approach presented in the literature to date. This innovation brings a notable impact in ERP-based communicators, appointing to the development of a faster and more reliable BCI technology.

Sensory Integration: Neuronal Filters for Polarized Light Patterns

Animal and human behaviour relies on local sensory signals that are often ambiguous. A new study shows how tuning neuronal responses to celestial cues helps locust navigation, demonstrating a common principle of sensory information processing: the use of matched filters.

Estimation of ultrasonic attenuation in a bone using coded excitation

This paper describes a novel approach to estimate broadband ultrasound attenuation (BUA) in a bone structure in human in vivo using coded excitation. BUA is an accepted indicator for assessment of osteoporosis. In the tested approach a coded acoustic signal is emitted and then the received echoes are compressed into brief, high amplitude pulses making use of matched filters and correlation receivers. In this way the acoustic peak pressure amplitude probing the tissue can be markedly decreased whereas the average transmitted intensity increases proportionally to the length of the code. This paper examines the properties of three different transmission schemes, based on Barker code, chirp and Golay code. The system designed is capable of generating 16 bits complementary Golay code (CGC), linear frequency modulated (LFM) chirp and 13-bit Barker code (BC) at 0.5 and 1 MHz center frequencies. Both in vivo data acquired from healthy heel bones and in vitro data obtained from human calcaneus were examined and the comparison between the results using coded excitation and two cycles sine burst is presented. It is shown that CGC system allows the effective range of frequencies employed in the measurement of broadband acoustic energy attenuation in the trabecular bone to be doubled in comparison to the standard 0.5 MHz pulse transmission. The algorithm used to calculate the pairs of Golay sequences of the different length, which provide the temporal side-lobe cancellation is also presented. Current efforts are focused on adapting the system developed for operation in pulse-echo mode; this would allow examination and diagnosis of bones with limited access such as hip bone.

Clustering to improve matched filter detection of weak gas plumes in hyperspectral thermal imagery

The use of matched filters on hyperspectral data has made it possible to detect faint signatures. This study uses a modified k-means clustering to improve matched filter performance. Several simple bivariate cases are examined in detail, and the interaction of filtering and partitioning is discussed. The authors show that clustering can reduce within-class variance and group pixels with similar correlation structures. Both of these features improve filter performance. The traditional k-means algorithm is modified to work with a sample of the image at each iteration and is tested against two hyperspectral datasets. A new "extreme" centroid initialization technique is introduced and shown to speed convergence. Several matched filtering formulations (the simple matched filter, the clutter matched filter, and the saturated matched filter) are compared for a variety of number of classes and synthetic hyperspectral images. The performance of the various clutter matched filter formulations is similar, all are about an order of magnitude better than the simple matched filter. Clustering is found to improve the performance of all matched filter formulations by a factor of two to five. Clustering in conjunction with clutter matched filtering can improve fifty-fold over the simple case, enabling very weak signals to be detected in hyperspectral images.

Comparison of transmission loss for underwater sound pulses measured using matched filters and integrated received power

Transmission loss (TL) may be found from the amplitude of a matched filter correlation applied to a pulsed acoustical signal. Use of matched filters to determine TL is advantageous because the technique can be applied continuously in real time without previously determining the pulse arrival time. However, distortion of the transmitted waveform by scattering, multiple path propagation, and Doppler frequency shift reduces the amplitude of the matched filter output thus increasing the TL determined in this way. The conventional TL calculation involves filtering the signal and integrating the received power for the duration of the pulse which yields a TL based on total received energy. This method must be applied after data is collected because the time of arrival of the pulses must first be determined. TL measurements derived using both methods were compared for data gathered during two trials in the Gulf of Mexico in both shallow and deep water environments. The transmitted waveforms were linear frequency-modulated pulses with a frequency range of 10 Hz, centered at 1055, 1225, and 1395 Hz. Results of the comparison will be presented.

Multiobject pattern recognition and detection in noisy backgrounds using a hierarchical approach

Matched filters are often used to detect image objects. However, for an image scene consisting of many different patterns corrupted by noise, the direct use of matched filters is time consuming and the performance of the matched filter deteriorates if the noise is not stationary and white. We develop a hierarchical approach for the detection of multiple objects which is divided into three steps: (a) prefiltering; (b) pattern recognition; and (c) object detection. This approach reduces computation time by more than 50% and increases classification efficiency compared with the direct matched filtering approach.

Spectral Dependance of Inverse Problem in Radiant Energy Transfer Through Haze

The spectral distribution of source intensity needed to obtain a prespecified irradiance at the receiver for different atmospheric haze condition has been calculated on the basis of semi-empirical expression due to Eldridge and Johnson (1962). Variations in the spectral distributions at the receiving end due to a general service lamp for different atmospheric condition are removable by use of matched filters in front of the source. For this study the scattering media are assumed to consist of uniform water dropl.ts of 0.1 (0.1) 0.6 micron radii with a concentration of 103 particles per c.c. The receiving distances are 0.25 (0.25) 1.0 (0.5) 4 0 Km. Attempt has been made to find out a suitable single matched filter for particular atmospheric haze condition to obtain uniform receiving illumination (with 5 percent variation) for all distances upto 1 Km.

Utilization of Matched Filters in Cochannel Interference Suppression

This paper discusses the use of matched filters in the suppression of cochannel interference. Attention is particularly focused on matched filters which exhibit a large time-bandwidth product (i.e., spread spectrum systems). For these cases, the output signal-to-interference power ratio is determined for any type of interfering signal. The central limit theorem also allows the determination of error probabilities for these cases. In the determination of error probabilities, two receiver configurations have been employed, one a completely linear receiver employing a simple threshold and the other a receiver which uses an envelope detector followed by a threshold decision mechanism. The results of a preliminary investigation of the low-time band-width case have also been discussed and show that smaller amounts of processing gain can also be obtained for this case. A tabulated summary of the gains available for 81 signal and interference cases has been given.

Matched Filter Detectors for Additive Arrays or Receivers near Reflectors

A matched filter operates on a signal to yield an output that is the autocorrelation of the input signal. To study the use of matched filters with an additive array, we did the following experiments: A loudspeaker was driven with an impulse, and the sound observed with an array of microphones. The electrical outputs of the microphones were added without time delays to yield a single channel of information. The matched filter-shift register delay line was adjusted to match the signal from the array when the array was in the end-fire direction. The array-matched filter combination had a correlation maximum for the direction in which the array output and filter were matched and the maximum decreased as the array was turned away from this direction. A single receiver near several reflectors also yields several arrivals on a single channel of information. The balance of the experiment was similar to the array-matched filter experiment. [Columbia University, Hudson Laboratories Informal Documentation No. 15. This work was supported by the Office of Naval Research under Contract Nonr-266(84).]