Recurrence Plots Research Articles

Research on detection of potato varieties based on spectral imaging analytical algorithm

Potatoes are popular among consumers due to their high yield and delicious taste. However, due to the numerous varieties of potatoes, different varieties are suitable for different processing methods. Therefore, it is necessary to distinguish varieties after harvest to meet the needs of processing enterprises and consumers. In this study, a new visible-near-infrared spectroscopic analysis method was proposed, which can achieve detection of five potato varieties. The method measures the transmission and reflection spectra of potatoes using a spectral acquisition system, encodes one-dimensional spectra into two-dimensional images using Gramian Angular Summation Field (GASF), Gramian Angular Difference Field (GADF), Markov Transition Field (MTF) and Recurrence Plot (RP), and improves the coordinated attention mechanism module and embeds the improved module into the ConvNeXt V2 model to build the ConvNeXt V2-CAP model for potato variety classification. The results show that compared with directly using one-dimensional classification models, image encoding of spectral data for classification greatly improves the accuracy. Among them, the best accuracy of 99.54% is achieved by using GADF image encoding of transmission spectra combined with the ConvNeXt V2-CAP model for classification, which is 16.28% higher than the highest accuracy of the one-dimensional classification model. The CAP attention mechanism module improves the performance of the model, especially when the dataset is small. When the training set is reduced to 150 images, the accuracy of the model is improved by 2.33% compared to the original model. Therefore, it is feasible to classify potato varieties using visible-near infrared spectroscopy and image encoding technology.

Nonlinear and nonstationary detection and quantification of multi-scale measured signals for bridge structure

The assessment of nonlinear and nonstationary levels in measured bridge signals is a vital step in system identification and long-term health monitoring for the bridge structure. The field-measured signals from the bridge structure are inherently weak and multiscale, so a specific adaptive variational mode decomposition (AVMD) is proposed to decompose them and extract their included multi-scale features. Combination the adaptability of empirical mode decomposition with the dimensionality reduction of principal component analysis, the number of inherent mode functions (IMFs) that need to be given in the conventional variational mode decomposition is adaptively determined in the proposed AVMD. The original measured signals from the bridge structure multiscale are subsequently decomposed by AVMD into the multiscale IMFs with the lowest cross-correlation. Then, the recurrence plot and recurrence quantification analysis are introduced into the detection and quantification of the measured signals, and the nonlinear and nonstationary quantification indexes are constructed to describe quantitatively the nonlinear and nonstationary levels. The stabilities and accuracies of three nonlinear and three nonstationary quantification indexes are comparatively discussed by the nonlinear and nonstationary detection and quantification of three well-defined simulated signals. The Shannon entropy and trapping time indexes are subsequently determined to quantify the nonlinear and nonstationary levels of the measured signals, respectively. Finally, the proposed algorithm and quantification indexes are applied to the nonlinear and nonstationary detection and quantification of the measured signals from the real-world bridge structures. It is shown from the validation and discussion that the proposed algorithm is available to detect and quantify the nonlinear and nonstationary levels of the measured multiscale signal from the real-world bridge structure.

Examining dependencies among different time scales in episodic memory - an experience sampling study.

We re-examined whether different time scales such as week, day of week, and hour of day are independently used during memory retrieval as has been previously argued (i.e., independence of scales). To overcome the limitations of previous studies, we used experience sampling technology to obtain test stimuli that have higher ecological validity. We also used pointwise mutual information to directly calculate the degree of dependency between time scales in a formal way. Participants were provided with a smartphone and were asked to wear it around their neck for two weeks, which was equipped with an app that automatically collected time, images, GPS, audio and accelerometry. After a one-week retention interval, participants were presented with an image that was captured during their data collection phase, and were tested on their memory of when the event happened (i.e., week, day of week, and hour). We find that, in contrast to previous arguments, memories of different time scales were not retrieved independently. Moreover, through rendering recurrence plots of the images that the participants collected, we provide evidence the dependency may have originated from the repetitive events that the participants encountered in their daily life.

Combustion instability analysis in an ethylene-fueled scramjet combustor under various fuel penetration height conditions using an image-based nonlinear dimensionality reduction method

The present study investigates the effects of fuel penetration height on combustion instabilities in an ethylene-fueled scramjet model combustor with a cavity flameholder. Experiments were performed at the stagnation temperature of 1900 K, the stagnation pressure of 0.37 MPa, and the Mach number of 2. Three fuel injection orifice diameters:2.3, 2.4, and 2.5 mm, were tested to elucidate the effects of ethylene penetration height on combustion instabilities. Furthermore, high-speed measurements of CH* chemiluminescence and shadowgraphs were performed with high-speed video cameras. To examine the dynamics of the combustion instabilities, time-resolved CH* chemiluminescence images and shock parameters, extracted from snapshots of shadowgraphs, were analyzed using a non-linear dimensionality reduction algorithm: Gaussian Process Dynamical Model (GPDM). Thereafter, further analyses on the acquired latent variables were conducted with Recurrence Plot (RP). The experimental results showed that the fuel equivalence ratio (ϕ) range for cavity shear-layer combustion mode expanded as d decreased. Furthermore, for ϕ = 0.18, the instability behavior remarkably changed at around d = 2.4 mm. Therefore, the instability dynamics for ϕ = 0.18 were investigated using GPDM and RP including results from our previous study (d = 2, 3, and 4 mm), revealing differences in the instability behaviors. For d = 3 and 4 mm, jet-wake combustion and ram combustion modes were established alternately with a frequency of about 1600 Hz. In contrast, at d = 2.4 and 2.5 mm, although a similar instability in the d = 4 mm case was present at almost the same oscillation frequency, a different instability behavior was also confirmed. This additional instability exhibited an intermediate state between jet-wake combustion and ram combustion modes. These two instabilities emerged aperiodically. For an unstable combustion in shear layer observed for d = 2 and 2.3 mm, corresponding RPs exhibited black patches, indicating that the oscillation amplitude diminished substantially.

Rolling Bearing Fault Diagnosis Based on Recurrence Plot

Rolling Bearing Fault Diagnosis Based on Recurrence Plot

Human activity recognition: an approach 2D CNN-LSTM to sequential image representation and processing of inertial sensor data

<p>The field of human activity recognition, abbreviated as HAR, benefits significantly from deep learning by addressing the complexity of human behavior and the vast volume of data produced by sensors. This work adopted the strategy of converting inertial data, such as accelerometer and gyroscope signals, into 2D images through recurrence plots. This approach facilitated the effective exploration of data input and neural network architectures. By utilizing the recent history of movements as input for the models, this study evaluated the impact of this methodology on HAR using two adapted architectures: 2D convolutional neural networks combined with long short-term memory layers (2D CNN-LSTM) and standalone 2D convolutional neural networks (2D CNN). Their performances were compared with other state-of-the-art deep learning models. The contributions of this study were threefold: the handling of input data, the development of the two network architectures for HAR, and the high accuracy achieved, ranging from 97% to 98%, on the public University of California, Irvine human activity recognition dataset (UCI-HAR). These results highlighted the benefit of incorporating temporal data to enhance accuracy in activity classification.</p>

EMG Physical Action Detection using Recurrence Plot Approach

Electromyography (EMG) is used to identify neuromuscular illnesses, motor issues, nerve damage, and degenerative ailments. EMG signals are difficult to accurately classify because of their complexity, nonlinearity, and time-variable nature; therefore, proper feature extraction and classification algorithms must be used. The proposed study uses the recurrence plot (RP) approach combined with deep learning (DL) techniques for the purpose of normal and aggressive EMG physical action (PA) classification. The transfer learning (TL) approach Inception-ResNet-v2 is utilized for feature extraction here. The Inception-ResNet-v2 model’s results demonstrate the excellent generalizability of the proposed method. Using only a few of the eight leads in the original EMG data, the model demonstrated the best accuracy (0.9841) and F1 score (0.99). These results show that the 2D RP-based technique has a significant clinical potential for classifying PA while requiring fewer EMG leads.

MEASURING STRUCTURAL CHANGES OF RECURRENCE PATTERNS IN MULTIFRACTAL AND MULTISCALE ASPECTS BY GENERALIZED RECURRENCE LACUNARITY

Recurrence lacunarity has been recently proposed to detect dynamical state transitions over various temporal scales. In this paper, we combine suggested distribution moments and introduce multifractal recurrence lacunarity to unearth rich information of trajectories in phase space. By considering generalized moments, it provides an enhanced measurement to account for differences of black pixels in the recurrence plot at various scales. Numerical simulations have proved that the proposed method is able to differentiate varying types of time series and provide further insights of inherent features including stochastic series, chaotic maps and series contaminated interference components. In real-world applications, it performs well on quantifying the subtle structural changes of financial time series. In addition, it is intriguing to confirm that corrugation signals possess much more vivid information of heterogeneity in terms of recurrence plots than normal ones.

Enhancing EEG-Based Emotion Recognition Using Asymmetric Windowing Recurrence Plots

Enhancing EEG-Based Emotion Recognition Using Asymmetric Windowing Recurrence Plots

Recurrence Plot Aided Partial Discharge Detection Framework Employing HFCT Sensor and Customized Convolutional Neural Network

Recurrence Plot Aided Partial Discharge Detection Framework Employing HFCT Sensor and Customized Convolutional Neural Network

Novel Parameter Error Identification Method for Power Plant Dynamic Models

Incorrect parameters in a power plant dynamic model affect the analysis results and control decisions in a power system, which may have serious consequences. The online calibration techniques in the literature utilize sensitivity analysis to determine the identifiable parameters before the calibration, and treat the majority of sensitive parameters as potentially erroneous. Hence, the candidate parameter subset for calibration mostly consists of “distractors,” i.e. highly sensitive but already correct parameters. The inclusion of distractors in the parameter calibration may decrease the accuracy of calibration, and increase the computation time. This paper proposes a method to identify erroneous parameters through the mismatch between collected PMU measurements and power plant model response. The proposed method relies on the classification of Recurrence Plots of the mismatch using 2-D Convolutional Neural Networks and identifies the erroneous parameters via the orthogonal decomposition. Hence, a smaller subset of candidate parameters is obtained by eliminating the distractors.

Statistical inference for microstate distribution in recurrence plots

Recurrence is an important property to unveil patterns in dynamical systems. The recurrence plot (RP) is a matrix that summarizes recurrence information from data sequences. Recently, a new approach to the RP have been developed to generalize the standard absence or occurrence of recurrences in the RP. The microstate concept was introduced to statistically characterize samples of any size in the RP. We explore in this paper the relation between microstates of different sizes in the RP. We distinguish two approaches: the top-bottom and the down-up analysis. In the first we have information from larger microstates to make inference about smaller ones, in the second the opposite. We establish analytical results that statistically estimate the microstate probabilities between different sample sizes. The analytical results for the normalized entropy were tested using two time series: a simple white noise case and a series from a logistic equation in the chaotic scenario. For both cases, the top-bottom and the down-up analytical relations were tested with good results. To depict the actual probability distributions, we used the Lorenz oscillator for both, the top-down and the bottom-up strategies, showcasing the limitations and perspectives for future works. We believe our results can open new theoretical paths in the RP analysis.

Research on fault diagnosis method of linear vibration screen based on fused RP-improved CNN

A fault diagnosis method based on fused RP (recurrence plot) and improved CNN (convolutional neural networks) is proposed for the traditional fault diagnosis method of linear vibrating screen manually designed and optimized features with feature quality uncertainty. A deep convolutional neural network combination model (MDCNN) with high-level feature fusion is designed. The collected multi-source vibration signals were converted into black and white recurrence plot, and fusion into three-channel RGB recurrence plot. Fault diagnosis was classified by the combination model of deep convolution neural network with high level feature fusion. The results show that the recurrence plot of homologous multi-sensor signals contains more information about the characteristics of the vibrating screen than the original colorful signal map or the Gram Point Field, which contributes to the feature learning and classification of MDCNN, with an average recognition accuracy of 97.59% under strong vibrating sift-5db noise. Compared with manifold classical neural network models, the MDCNN model is proved to be effective, stable and small space occupying.

A novel approach for sports injury risk prediction: based on time-series image encoding and deep learning.

The rapid development of big data technology and artificial intelligence has provided a new perspective on sports injury prevention. Although data-driven algorithms have achieved some valuable results in the field of sports injury risk assessment, the lack of sufficient generalization of models and the inability to automate feature extraction have made it challenging to deploy research results in the real world. Therefore, this study attempts to build an injury risk prediction model using a combination of time-series image encoding and deep learning algorithms to address this issue better. This study used the time-series image encoding approach for feature construction to represent relationships between values at different moments, including Gramian Angular Summation Field (GASF), Gramian Angular Difference Field (GADF), Markov Transition Field (MTF), and Recurrence Plot (RP). Deep Convolutional Auto-Encoder (DCAE) learned the image-encoded data for representation to obtain features with good discrimination, and the classifier was performed using Deep Neural Network (DNN). The results from five repeated experiments show that the GASF-DCAE-DNN model is overall better in the training (AUC: 0.985 ± 0.001, Gmean: 0.930 ± 0.007, Sensitivity: 0.997 ± 0.003, Specificity: 0.868 ± 0.013) and test sets (AUC: 0.891 ± 0.026, Gmean: 0.830 ± 0.027, Sensitivity: 0.816 ± 0.039, Specificity: 0.845 ± 0.022), with good discriminative power, robustness, and generalization ability. Compared with the best model reported in the literature, the AUC, Gmean, Sensitivity, and Specificity of the GASF-DCAE-DNN model were higher by 23.9%, 27.5%, 39.7%, and 16.2%, respectively, which confirmed the validity and practicability of the model in injury risk prediction. In addition, differences in injury risk patterns between the training and test sets were identified through shapley additivity interpretation. It was also found that the training volume was an essential factor that affected injury risk prediction. The model proposed in this study provides a powerful injury risk prediction tool for future sports injury prevention practice.

Improvement of automatic speech recognition systems utilizing 2D adaptive wavelet transformation applied to recurrence plot of speech trajectories

Improvement of automatic speech recognition systems utilizing 2D adaptive wavelet transformation applied to recurrence plot of speech trajectories

A new similarity measurement method for time series based on image fusion of recurrence plots and wavelet scalogram

The core of data series classification and clustering is how to evaluate the similarity between time series. However, for one-dimensional time series, the vast majority of similarity measurement methods currently relied on one-dimensional information. Although some classifications of time series were achieved by the images transformed from time series, they mostly relied on a single imaging mechanism (time domain or frequency domain). In order to give a new idea for time series similarity measurement, this paper proposed a similarity measurement method of time series based on the fused images. According to recurrence plots (RP) and wavelet scalogram (WS), the original time series was first transformed into images. Then, RP and WS were fused into an image. Next, the feature in the image was extracted by ResNet-18. Finally, the similarity of the time series can be obtained by the Euclidean distance between the features extracted by ResNet-18. The proposed method was verified by the time series from UCR Time Series Classification/Clustering Homepage. The results showed that the similarity measurement of time series can be effectively improved by fusing the images obtained by different imaging mechanisms in the time domain and frequency domain.

Fault Diagnosis Method for Human Coexistence Robots Based on Convolutional Neural Networks Using Time-Series Data Generation and Image Encoding.

This paper proposes fault diagnosis methods aimed at proactively preventing potential safety issues in robot systems, particularly human coexistence robots (HCRs) used in industrial environments. The data were collected from durability tests of the driving module for HCRs, gathering time-series vibration data until the module failed. In this study, to apply classification methods in the absence of post-failure data, the initial 50% of the collected data were designated as the normal section, and the data from the 10 h immediately preceding the failure were selected as the fault section. To generate additional data for the limited fault dataset, the Wasserstein generative adversarial networks with gradient penalty (WGAN-GP) model was utilized and residual connections were added to the generator to maintain the basic structure while preventing the loss of key features of the data. Considering that the performance of image encoding techniques varies depending on the dataset type, this study applied and compared five image encoding methods and four CNN models to facilitate the selection of the most suitable algorithm. The time-series data were converted into image data using image encoding techniques including recurrence plot, Gramian angular field, Markov transition field, spectrogram, and scalogram. These images were then applied to CNN models, including VGGNet, GoogleNet, ResNet, and DenseNet, to calculate the accuracy of fault diagnosis and compare the performance of each model. The experimental results demonstrated significant improvements in diagnostic accuracy when employing the WGAN-GP model to generate fault data, and among the image encoding techniques and convolutional neural network models, spectrogram and DenseNet exhibited superior performance, respectively.

Large eddy simulation of combustion instabilities in a co-axial stratified model combustor under transient operating conditions

This work conducted large eddy simulation investigation on the effect of transient time and transient direction on the combustion instabilities in a co-axial stratified model combustor under transient operating conditions. Two transient times, which is greater than and less than the time delay, respectively, 0.1 s and 0.1 ms, were studied in different transient direction. The transient combustion process is analyzed by the global HRR, temperature, CH and CO. The time–frequency analysis of the pilot-stage and main-stage are carried out by FFT and SWT. The mode transition characteristics of the pilot-stage and main-stage are discussed based on the recurrence plots and Poincare plots. Finally, the bifurcations of the HRR and temperature caused by the transient time are analyzed. It was found that the shorter transient time is associated with larger oscillation amplitude of temperature, as well as the greater temperature and HRR gradient, and the maximum oscillation amplitude is larger. Furthermore, due to the shorter transient time in case ⅠⅡ-, the changing rate of heat release rate is approximately 725 J·m−3·s−2, which is significantly greater than the 72 J·m−3·s−2 in case ⅠⅡ+. In case ⅠⅡ-, the maximum CH is 2.0e-12, which is approximately 11 % higher than the maximum value in other cases. Different modes were observed in different transient time and transient direction. Longer transient time is associated with both a greater number of modes and frequency of mode transitions, and higher possibility of excitation oscillation occurring. Under a specific global average mixture fraction of 0.0345, two different flame shapes were observed in separate cases: a stratified flame and an M-shaped flame.

Comparison of electrogram characteristics in persistent atrial fibrillation.

Multiple analysis techniques evaluate electrograms during atrial fibrillation (AF), but none have been established to guide catheter ablation. This study compares electrogram properties recorded from multiple right (RA) and left atrial (LA) sites. Multisite LA/RA mapping (281 ± 176/239 ± 166 sites/patient) was performed in 42 patients (30 males, age 63 ± 9 years) undergoing first (n = 32) or redo-AF ablation (n = 10). All electrogram recordings were visually reviewed and artifactual signals were excluded leaving a total of 21 846 for analysis. Electrogram characteristics evaluated werecycle length (CL), amplitude, Shannon's entropy (ShEn), fractionation interval, dominant frequency, organizational index, and cycle length of most recurrent morphology (CLR ) from morphology recurrence plot analysis. Electrogram characteristics were correlated to each other. All pairwise comparisons were significant (p < .001) except for dominant frequency and CLR (p = .59), and amplitude and dominant frequency (p = .38). Only ShEn and fractionation interval demonstrated a strong negative correlation (r = -.94). All other pairwise comparisons were poor to moderately correlated. The relationships are highly conserved among patients, in the RA versus LA, and in those undergoing initial versus redo ablations. Antiarrhythmic drug therapy did not have a significant effect on electrogram characteristics, except minimum ShEn. Electrogram characteristics associated with ablation outcome were shorter minimum CLR , lower minimum ShEn, and longer mimimum CL. There was minimal overlap between the top 10 sites identified by one electrogram characteristic and the top 10 sites identified by the other 10 characteristics. Multiple techniques can be employed for electrogram analysis in AF. In this analysis of eightdifferent electrogram characteristics, seven were poorly to moderately correlated and do not identify similar locations. Only some characteristics were predictive of ablation outcome. Further studies to consider electrogram properties, perhaps in combination, for categorizing and/or mapping AF are warranted.

Correlated spectral and recurrence variations of Cygnus X-1

ABSTRACT We present results of recurrence analysis of the black hole X-ray binary Cygnus X-1 using combined observations from the Rossi X-ray Timing Explorer All-sky Monitor and the Japanese Monitor of All-sky X-ray Image aboard the International Space Station. From the time-dependent windowed recurrence plot (RP), we compute 10 recurrence quantities that describe the dynamical behaviour of the source and compare them to the spectral state at each point in time. We identify epochs of state changes corresponding to transitions into highly deterministic or highly stochastic dynamical regimes and their correlation to specific spectral states. We compare k-Nearest Neighbors and Random Forest models for various sizes of the time-dependent RP. The spectral state in Cygnus X-1 can be predicted with greater than 95 per cent accuracy for both types of models explored across a range of RP sizes based solely on the recurrence properties. The primary features from the RP that distinguish between spectral states are the determinism, Shannon entropy, and average line length, all of which are systematically higher in the hard state compared to the soft state. Our results suggest that the hard and soft states of Cygnus X-1 exhibit distinct dynamical variability and the time domain alone can be used for spectral state classification.