Adaptive Multiscale Entropy Research Articles

Adaptive crack damage identification based on multi-scale sample entropy under variable temperature environment

High-speed trains inevitably experience the impact of complex temperature loads during operation, which can easily result in cracks. Structural health monitoring methods based on Lamb waves are affected by the impact of temperature on their propagation mechanism, leading to difficulties in constructing accurate damage diagnostic models. This paper proposes a crack damage identification method based on adaptive multi-scale sample entropy under variable temperature environment. A sliding window method is proposed to obtain multiple wave packets in a Lamb wave, which avoids the disadvantage of insufficient information of a single wave packet and improves the robustness and reliability of diagnosis. A variance-based multiscale transform method is proposed to process Lamb wave signals, which reduces the sensitivity of Lamb wave signals to temperature compared to the traditional mean-based multiscale transform method. Lamb waves in different sliding windows are transformed at multiple scales to extract damage Multi-scale sample entropy (MSE) at different scales. The Quantum genetic algorithm (QGA) is introduced to optimize sliding windows and MSE, achieving adaptive MSE extraction under variable temperature environments. Based on the advantage of adaptive multiscale entropy, a quantitative crack diagnosis model is established. To verify the effectiveness of the proposed method, crack detection experiments under variable temperatures were conducted. The results show that the proposed adaptive MSE has good resistance to temperature changes and can accurately identify crack length.

Research on fatigue driving detection using forehead EEG based on adaptive multi-scale entropy

Fatigue driving is one of the main factors that causes traffic accidents. In the current non-linear analysis methods, the entropy feature extraction methods can be well applied to detection of driving fatigue. However, all of these methods analyzed the EEG data on a single scale signal and there is also no effective way to determine the signal multi-scale information. In addition, most of the current researches choose all the electrodes, which is not conducive to practical application. Based on the forehead EEG data, an adaptive multi-scale entropy feature extraction algorithm is proposed by combining with an adaptive scaling factor (ASF) obtaining algorithm and entropy feature extraction method. Firstly, ASF algorithm is used to extract the scale factor of the signal. Secondly, this factor is used to reconstruct the signal to get new signal data. Finally, the entropy features are extracted for classification. The experimental results show that the proposed adaptive multi-scale entropy feature algorithm is effective in the detection of fatigue driving based on using forehead EEG data. So the effectiveness of this feature extraction algorithm is proved.

Adaptive Multi-Scale Entropy Fusion De-Hazing Based on Fractional Order

This paper describes a proposed fractional filter-based multi-scale underwater and hazy image enhancement algorithm. The proposed system combines a modified global contrast operator with fractional order-based multi-scale filters used to generate several images, which are fused based on entropy and standard deviation. The multi-scale-global enhancement technique enables fully adaptive and controlled color correction and contrast enhancement without over exposure of highlights when processing hazy and underwater images. This in addition to the illumination/reflectance estimation coupled with global and local contrast enhancement. The proposed algorithm is also compared with the most recent available state-of-the-art multi-scale fusion de-hazing algorithm. Experimental comparisons indicate that the proposed approach yields a better edge and contrast enhancement results without a halo effect, without color degradation, and is faster and more adaptive than all other algorithms from the literature.

PERCEPTUAL SUPPRESSION REVEALED BY ADAPTIVE MULTI-SCALE ENTROPY ANALYSIS OF LOCAL FIELD POTENTIAL IN MONKEY VISUAL CORTEX

Generalized flash suppression (GFS), in which a salient visual stimulus can be rendered invisible despite continuous retinal input, provides a rare opportunity to directly study the neural mechanism of visual perception. Previous work based on linear methods, such as spectral analysis, on local field potential (LFP) during GFS has shown that the LFP power at distinctive frequency bands are differentially modulated by perceptual suppression. Yet, the linear method alone may be insufficient for the full assessment of neural dynamic due to the fundamentally nonlinear nature of neural signals. In this study, we set forth to analyze the LFP data collected from multiple visual areas in V1, V2 and V4 of macaque monkeys while performing the GFS task using a nonlinear method - adaptive multi-scale entropy (AME) - to reveal the neural dynamic of perceptual suppression. In addition, we propose a new cross-entropy measure at multiple scales, namely adaptive multi-scale cross-entropy (AMCE), to assess the nonlinear functional connectivity between two cortical areas. We show that: (1) multi-scale entropy exhibits percept-related changes in all three areas, with higher entropy observed during perceptual suppression; (2) the magnitude of the perception-related entropy changes increases systematically over successive hierarchical stages (i.e. from lower areas V1 to V2, up to higher area V4); and (3) cross-entropy between any two cortical areas reveals higher degree of asynchrony or dissimilarity during perceptual suppression, indicating a decreased functional connectivity between cortical areas. These results, taken together, suggest that perceptual suppression is related to a reduced functional connectivity and increased uncertainty of neural responses, and the modulation of perceptual suppression is more effective at higher visual cortical areas. AME is demonstrated to be a useful technique in revealing the underlying dynamic of nonlinear/nonstationary neural signal.

Adaptive Multiscale Entropy Analysis of Multivariate Neural Data

Multiscale entropy (MSE) has been widely used to quantify a system's complexity by taking into account the multiple time scales inherent in physiologic time series. The method, however, is biased toward the coarse scale, i.e., low-frequency components due to the progressive smoothing operations. In addition, the algorithm for extracting the different scales is not well adapted to nonlinear/nonstationary signals. In this letter, we introduce adaptive multiscale entropy (AME) measures in which the scales are adaptively derived directly from the data by virtue of recently developed multivariate empirical mode decomposition. Depending on the consecutive removal of low-frequency or high-frequency components, our AME can be estimated at either coarse-to-fine or fine-to-coarse scales over which the sample entropy is performed. Computer simulations are performed to verify the effectiveness of AME for analysis of the highly nonstationary data. Local field potentials collected from the visual cortex of macaque monkey while performing a generalized flash suppression task are used as an example to demonstrate the usefulness of our AME approach to reveal the underlying dynamics in complex neural data.