Controlled-source electromagnetic (CSEM) data recorded in industrialized areas are inevitably contaminated by strong cultural noise. Traditional noise attenuation methods are often ineffective for intricate aperiodic ...
详细信息
Controlled-source electromagnetic (CSEM) data recorded in industrialized areas are inevitably contaminated by strong cultural noise. Traditional noise attenuation methods are often ineffective for intricate aperiodic noise. To address the above mentioned problem, we have developed a novel noise isolation method based on the fast Fourier transform, complementary ensemble empirical mode decomposition (CEEMD), and shiftinvariantsparsecoding (SISC, an unsupervised machine-learning algorithm under a data-driven framework). First, large powerline noise is accurately subtracted in the frequency domain. Then, the CEEMD-based algorithm is used to correct the large baseline drift. Finally, taking advantage of the sparsity of periodic signals, SISC is applied to autonomously learn a feature atom (the useful signal with a length of one period) from the detrended signal and recover the CSEM signal with high accuracy. We determine the performance of the SISC by comparing it with three other promising signal processing methods, such as the mathematic morphology filtering, soft-threshold wavelet filtering, and K-singular-value decomposition (another dictionary learning method) sparse decomposition. Experimental results illustrate that SISC provides the best performance. Robustness test results indicate that SISC can increase the signal-to-noise ratio of noisy signal from 0 to more than 15 dB. Case studies of synthetic and real data collected in the Chinese provinces of Sichuan and Yunnan indicate that our method is capable of effectively recovering the useful signal from the observed data contaminated with different kinds of strong ambient noise. The curves of U/I and apparent resistivity after applying our method improved greatly. Moreover, our method performs better than the robust estimation method based on correlation analysis.
In machinery fault diagnosis, it is common that one kind of fault may correspond to several conditions, these conditions may contain different loads, different speeds and so on. When using conventional intelligent mac...
详细信息
In machinery fault diagnosis, it is common that one kind of fault may correspond to several conditions, these conditions may contain different loads, different speeds and so on. When using conventional intelligent machinery fault diagnosis methods on diagnosing this kind of faults, if only one condition among all of these conditions was trained, when using this trained classifier for diagnosing fault which containing all conditions, it would obtain a classification result with higher error, it is the problem of robustness;but if we train all these data in each condition, the robustness can be improved a lot, but the time would be wasted. In order to balance these two aspects of fault diagnosis which seem contradict with each other, someone proposed a new method which based on shift-invariant sparse coding (SISC) method, this method can learn features from each condition of the same fault, and these features are adaptive to other conditions, which solve the first problem, but time efficiency of this algorithm is lower, in this paper, by improving the efficiency of shift-invariant sparse coding, we can reduce a lot of time on learning features. Through the experiment testing, it showed that this new method proposed in this paper produced better performance than SISC algorithm.
This paper is dedicated to feature extraction of acoustic emission (AE) signals from machining hard brittle materials. A diamond intender is used to scratch BK7 and sapphire. Typical characteristics of AE signals from...
详细信息
ISBN:
(数字)9781510623279
ISBN:
(纸本)9781510623279
This paper is dedicated to feature extraction of acoustic emission (AE) signals from machining hard brittle materials. A diamond intender is used to scratch BK7 and sapphire. Typical characteristics of AE signals from the two materials of are all concentrated in [100-200]KHz frequency span, and band-filtered signals represent obvious local burst-type waveforms, which is closely related to the crack and fragmentation phenomena that occur when brittle material is removed. Locations and oscillation form of burst-type AE are useful information for process monitoring and quality prediction of machined surface. In order to acquire the information, the theory of shift-invariant sparse coding(SISC) is introduced to analysis AE RMS signals. Experimental results of the two typical hard brittle materials show that oscillation form and locations of burst-type AE can be sparsely expressed by a self-learned atom and the corresponding coefficients. From the aspects of machining process monitoring, taking burst-type AE event as a monitor parameter is more accurate and objective in the reflection of locations and scales of cracks on the machined surface induced by machining.
Convolutional sparsecoding is an interesting alternative to standard sparsecoding in modeling shiftinvariant signals, giving impressive results for example in unsupervised learning of visual features. In state-of-t...
详细信息
Convolutional sparsecoding is an interesting alternative to standard sparsecoding in modeling shiftinvariant signals, giving impressive results for example in unsupervised learning of visual features. In state-of-the-art methods, the most time-consuming parts include inversion of a linear operator related to convolution. In this article we show how these inversions can be computed non-iteratively in the Fourier domain using the matrix inversion lemma. This greatly speeds up computation and makes convolutional sparsecoding computationally feasible even for large problems. The algorithm is derived in three variants, one of them especially suitable for parallel implementation. We demonstrate algorithms on two-dimensional image data but all results hold for signals of arbitrary dimension. (C) 2016 Elsevier Inc. All rights reserved.
In the signal processing domain, there has been growing interest in sparsecoding with a learned dictionary instead of a predefined one, which is advocated as an effective mathematical description for the underlying p...
详细信息
In the signal processing domain, there has been growing interest in sparsecoding with a learned dictionary instead of a predefined one, which is advocated as an effective mathematical description for the underlying principle of mammalian sensory systems in processing information. In this paper, sparsecoding is introduced as a feature extraction technique for machinery fault diagnosis and an adaptive feature extraction scheme is proposed based on it. The two core problems of sparsecoding, i.e., dictionary learning and coefficients solving, are discussed in detail. A natural extension of sparsecoding, shift-invariant sparse coding, is also introduced. Then, the vibration signals of rolling element bearings are taken as the target signals to verify the proposed scheme, and shift-invariant sparse coding is used for vibration analysis. With the purpose of diagnosing the different fault conditions of bearings, features are extracted following the proposed scheme: basis functions are separately learned from each class of vibration signals trying to capture the defective impulses;a redundant dictionary is built by merging all the learned basis functions;based on the redundant dictionary, the diagnostic information is made explicit in the solved sparse representations of vibration signals;sparse features are formulated in terms of activations of atoms. The multiclass linear discriminant analysis (LDA) classifier is used to test the discriminability of the extracted sparse features and the adaptability of the learned atoms. The experiments show that sparsecoding is an effective feature extraction technique for machinery fault diagnosis. (C) 2010 Elsevier Ltd. All rights reserved.
We present a framework for unsupervised detection of nonverbal behavioral cues-hand gestures, pose, body move merits, etc.-from a collection of motion capture (MoCap) sequences in a public speaking setting. We extract...
详细信息
ISBN:
(纸本)9781450334594
We present a framework for unsupervised detection of nonverbal behavioral cues-hand gestures, pose, body move merits, etc.-from a collection of motion capture (MoCap) sequences in a public speaking setting. We extract the cues by solving a sparse and shift-invariant dictionary learning problem, known as shift-invariant sparse coding. We find that the extracted behavioral cues are human-interpretable in the context of public speaking. Our technique can be applied to automatically identify the common patterns of body movements and the time-instances of their occurrences, minimizing time and efforts needed for manual detection and coding of nonverbal human behaviors.
暂无评论