Continuous Wavelet Transform Method Research Articles

A novel moisture damage detection method for asphalt pavement from GPR signal with CWT and CNN

To locate the moisture damage area in asphalt pavement, an automatic detection method by leveraging the state-of-the-art deep learning approach and continuous wavelet transform (CWT) method from high frequency GPR signal was presented. A numeric simulation and a laboratory experiment were employed to analyze the response of GPR signal in moisture damage area. GPR signals were transferred to time-frequency domain by using a CWT method, before comparing CWT results among normal pavement, moisture and bridge joints. Moisture damage datasets, including 8215 traces of moisture damages, 8215 traces of normal pavement and 7643 traces of bridge joints, were extracted from GPR field survey of bridge pavement with 2.3 GHz antenna. The maximum and mean values of CWT results were obtained from the dataset and used to normalize the CWT dataset. Typical classification models, including Resnet50, ResNet18, ShuffleNet, EfficientNet, NasNetmobile and DenseNet, were trained and compared on the CWT dataset. The result indicates that the CWT features among normal pavement, moisture damage and bridge joints have significant differences. Moreover, the energy distribution in the CWT image can represent the moisture damage depth and width information, proving that the CWT method is effective for extracting moisture damage features from GPR signal. The overall performance of ResNet50 model is best with high accuracy and high efficiency in moisture detection compared to other models. Experimental results proved the promising performance of our approach in detecting moisture damage in asphalt pavement, and this method could also be used for subsurface target detection in other GPR application.

Vehicle Response-Based Bridge Modal Identification Using Different Time–Frequency Analysis Methods

Modal identification of bridges based on the responses of moving vehicles, termed as the indirect method, has drawn much attention in recent years. This indirect method has been proven to be capable of identifying bridge frequencies and mode shapes in the feasibility research. However, in the presence of additional factors such as vehicle speed and road surface roughness, it is difficult to identify high-order frequencies and high-resolution mode shapes of the bridge. The authors previously proposed a tractor-double-trailers model and the time–domain subtraction method to obtain the relative displacement of two trailers for bridge modal identification. In this study, the authors develop a methodology to improve the identification performance by combining the EMD technique and the time–frequency analysis method. The EMD technique is applied to separate the multi-source response signals of the trailers to obtain the signal component that contains only the bridge vibration. Then, different time–frequency analysis methods such as the short-time Fourier transform (STFT), Wigner–Ville distribution (WVD), and continuous wavelet transform (CWT) are adopted to construct the high-resolution bridge mode shapes. Numerical studies with one-dimensional and three-dimensional bridge models are conducted to verify the proposed method and investigate the effects of different factors. Finally, a laboratory test is implemented on a scaled bridge to verify the performance of the proposed method. The results show that the proposed method is able to separate the vehicle frequency and bridge frequencies, which facilitates the identification of bridge frequencies from the dynamic response of moving vehicles. Meanwhile, the CWT method performs the best in the identification of modal shapes among the three time–frequency analysis methods.

The analysis on groundwater storage variations from GRACE/GRACE-FO in recent 20 years driven by influencing factors and prediction in Shandong Province, China

Monitoring and predicting the regional groundwater storage (GWS) fluctuation is an essential support for effectively managing water resources. Therefore, taking Shandong Province as an example, the data from Gravity Recovery and Climate Experiment (GRACE) and GRACE Follow-On (GRACE-FO) is used to invert GWS fluctuation from January 2003 to December 2022 together with Watergap Global Hydrological Model (WGHM), in-situ groundwater volume and level data. The spatio-temporal characteristics are decomposed using Independent Components Analysis (ICA), and the impact factors, such as precipitation and human activities, which are also analyzed. To predict the short-time changes of GWS, the Support Vector Machines (SVM) is adopted together with three commonly used methods Long Short-Term Memory (LSTM), Singular Spectrum Analysis (SSA), Auto-Regressive Moving Average Model (ARMA), as the comparison. The results show that: (1) The loss intensity of western GWS is significantly greater than those in coastal areas. From 2003 to 2006, GWS increased sharply; during 2007 to 2014, there exists a loss rate − 5.80 ± 2.28 mm/a of GWS; the linear trend of GWS change is − 5.39 ± 3.65 mm/a from 2015 to 2022, may be mainly due to the effect of South-to-North Water Diversion Project. The correlation coefficient between GRACE and WGHM is 0.67, which is consistent with in-situ groundwater volume and level. (2) The GWS has higher positive correlation with monthly Global Precipitation Climatology Project (GPCP) considering time delay after moving average, which has the similar energy spectrum depending on Continuous Wavelet Transform (CWT) method. In addition, the influencing facotrs on annual GWS fluctuation are analyzed, the correlation coefficient between GWS and in-situ data including the consumption of groundwater mining, farmland irrigation is 0.80, 0.71, respectively. (3) For the GWS prediction, SVM method is adopted to analyze, three training samples with 180, 204 and 228 months are established with the goodness-of-fit all higher than 0.97. The correlation coefficients are 0.56, 0.75, 0.68; RMSE is 5.26, 4.42, 5.65 mm; NSE is 0.28, 0.43, 0.36, respectively. The performance of SVM model is better than the other methods for the short-term prediction.

Cement pavement void detection algorithm based on GPR signal and continuous wavelet transform method

The dimension of the void area in pavement is crucial to its structural safety. However, there is no effective method to measure its geometric parameters. To address this issue, a void size extraction algorithm based on the continuous wavelet transform (CWT) method was proposed using ground-penetrating radar (GPR) signal. Firstly, the finite-difference time-domain (FDTD) method was used to investigate the GPR response of void areas with different shapes, sizes, and depths. Next, the GPR signal was processed using the CWT method, and a 3D image based on the CWT result was used to visualize the void area. Based on the differences between the void and normal pavement in the time and frequency domains, the signal with maximum energy from the CWT time–frequency result was extracted and combined to reconstruct the new B-scan image, where void areas have energy concentration phenomenon. Based on this, width and depth of void detection algorithm was proposed to recognize the void area. Finally, the detection algorithm was verified both in numerical model and physical lab model. The results indicated that the CWT time–frequency energy spectrum can be used to enhance the void feature, and the 3D CWT image can clearly visualize the void area with a highlighted energy area. After fully testing and validating in numerical and lab models, our proposed method achieved high accuracy in void width and depth detection, providing a precise method for estimating void dimension in pavement. This method can guide DOT departments to carry out pre-maintenance, thereby ensuring pavement safety.

Imaging Subsurface Channels within the Tertiary Niger Delta Basin using Spectral Decomposition Methods

Abstract Channels are hydrocarbon prospective areas which are often below the resolution of seismic data. This study was therefore conducted with the aim of using spectral decomposition methods for imaging of buried channels and associated structures in a 3D Post-Stack Depth Migrated (PSDM) volume acquired from a field in the Niger Delta Basin. Two commonly used spectral decomposition methods were used to enhance the resolution of seismic data for interpretation. The methods utilized included Fast Fourier Transform (FFT) and Continuous Wavelet Transform (CWT). Petrel and Opendtect were the interpretational software tools utilized for the study. The seismic data was conditioned using structural smoothing, and a zone of interest was selected around -1000 to -3000ms. Three dominant frequencies were selected from the sub-volume generated around the zone of interest. The frequencies were 10Hz, 25Hz and 32Hz respectively. These frequencies were extracted from the seismic data using FFT and CWT decomposition methods. The results were colored using Red, Green and Blue (Red for 10Hz, Green for 25Hz and Blue for 32Hz). These frequencies were then blended together in a process called RGB blending. The resultant output was a mix of only these three dominant frequencies. Analysis of these results revealed the presence of several channels, crevasse splay deposits, flood-plains and faults on two time-slices obtained at -2020ms and -1800ms. The channels were meandering channels of low sinuosity, some of which were fault controlled. The crevasse splay deposits were significantly large and increased in number at the shallow time-slice. Some of the crevasse splay deposits were found bifurcating fault lines and being deposited on floodplains, indicating that they were younger than the faults. Channel flow direction was from East to West, while channel migration was from South to North of the study area. The increase in the number of channels from the base to the top of the seismic data suggested a multistory stacked channel system. It was established that channels and crevasse splays were best enhanced using the FFT decomposition method while major and minor faults were best enhanced using CWT method. Hence, this study has proven that the spectral decomposition method is very effective in imaging subsurface buried channels, thus delineating prospects and can be used for reservoir characterization. Keywords: Spectral decomposition, Fast Fourier Transform (FFT), Continuous Wavelet Transform (CWT), Channels, Crevasse splays, Faults, Reservoir Characterization.

Frequency characteristics analysis of pipe-to-soil potential under metro stray current interference using continuous wavelet transform method

This study aims to investigate the impact of stray current interference from metro systems on the pipe-to-soil potentials of metallic pipelines, which is closely associated with stray current corrosion. The pipe-to-soil potentials were assessed using coupon testing, while in-situ measurement was conducted to measure the metro leakage currents. The collected field data from the pipe-to-soil potentials and metro leakage current measurements were subjected to a statistical analysis using Continuous Wavelet Transform (CWT) and linear correlation methods. The analysis revealed consistent dominant frequencies of interference at the same test point for both on-potentials and off-potentials conditions, as well as within different time periods. However, variations in dominant frequencies of interference were observed among test points along different sections of the pipeline. Notably, three primary test points exhibited frequencies of 8.3, 8.9, and 7.8 mHz, corresponding to the main interstation travel intervals of neighboring metro lines. Furthermore, the dominant period ranges of interference of metro stray current interference were found to align with the interstation travel interval ranges of the metro lines. When the OVPD acts, the leakage current of the metro station grounding system near the test point has a significant contribution to the amplitude of pipe-to-soil potential fluctuation. These findings highlight the significant role of interstation travel intervals in the periodicity of metro stray current interference. The identified dominant interference frequency ranges can be utilized to determine interference sources and evaluate the extent of metro stray current corrosion.

A Simple UV Spectrophotometric Method Based on Continuous Wavelet Transform and Multivariate Calibration Model for the Concurrent Analysis of Three Water-Soluble Vitamins in Fertility Supplements.

Owing to the presence of overlapping spectra in pharmaceutical components, classical spectrophotometry is hard for concurrent determination. The advance of chemometrics along with UV-Vis spectrophotometry has contributed to solving this problem. In this study, a fast, easy, precise, accurate, low-cost, and eco-friendly spectrophotometric technique was introduced and validated for the simultaneous analysis of vitamin B6, vitamin B12, and vitamin C in fertility supplements for men and women using continuous wavelet transform (CWT) and partial least squares (PLS) techniques without using time-consuming extraction process and organic solvents. In the CWT method, the zero-crossing technique was applied to obtain the optimum points for plotting calibration curves for each component. The validation of both methods was evaluated by analyzing several mixtures with different concentrations. The efficiency of the proposed methods was also surveyed on commercial capsules. Wavelet families, including Symlet (sym2) at 230, Biorthogonal (bior1.3) at 378 nm, and Daubechies (db2) at 261, were considered for vitamins B6, B12, and C, respectively. The linear range was found to be 8-20, 8-20, and 10-25 μg/mL with the coefficient of determination (R2) equal to 0.9982, 0.9978, and 0.9701 for B6, B12, and C, respectively. Low limit of detection (LOD) (<0.09 μg/mL) and limit of quantification (LOQ) <0.9 μg/mL were achieved. The mean recovery values in synthetic mixtures were from 98.38 to 98.89% and from 99.83 to 99.99%, where root-mean-square error (RMSE) of not more than 0.4 and 0.05 using the CWT and PLS methods, respectively. The obtained results from the commercial capsules, applying the suggested techniques, were compared to those yielded by the high-performance liquid chromatography (HPLC) method using the analysis of variance (ANOVA) test. According to the results, there are no significant differences, and they were in good agreement. According to all the mentioned cases, the proposed approaches can replace the time-consuming and costly HPLC method in quality control laboratories. Green spectrophotometry coupling chemometrics methods were proposed. Simultaneous determination of three water-soluble vitamins in fertility supplements was done using these approaches. Rapidity, simplicity, low cost, and accuracy are the benefits of the proposed methods. A HPLC technique was used as a reference method to compare with the chemometrics methods.

Water Surface Acoustic Wave Detection by a Millimeter Wave Radar

Feature extraction and recognition of underwater targets are important in military and civilian areas. This paper studied water surface acoustic wave (WSAW) detection by a millimeter wave (mmWave) radar. The mmWave-based endpoint detection method of the WSAW was introduced. Simulated results show that the continuous wavelet transform (CWT) method has a better detection performance. A 77 GHz large aperture antenna mmWave radar sensor and an underwater acoustic transmitter have been applied to conduct laboratory experiments. Still water surface experimental results verify that the CWT method has better detection capability, and the mmWave radar can accurately detect even 155 nm WSAW. Wavy water surface experimental results demonstrate the ability of the mmWave radar to analyze the time-frequency feature of the weak WSAW signal. These works indicate the potential of mmWave radar for the cross-medium detection and recognition of underwater targets.

A New Hybrid Approach Based on Time Frequency Images and Deep Learning Methods for Diagnosis of Migraine Disease and Investigation of Stimulus Effect.

Migraine is a neurological disorder that is associated with severe headaches and seriously affects the lives of patients. Diagnosing Migraine Disease (MD) can be laborious and time-consuming for specialists. For this reason, systems that can assist specialists in the early diagnosis of MD are important. Although migraine is one of the most common neurological diseases, there are very few studies on the diagnosis of MD, especially electroencephalogram (EEG)-and deep learning (DL)-based studies. For this reason, in this study, a new system has been proposed for the early diagnosis of EEG- and DL-based MD. In the proposed study, EEG signals obtained from the resting state (R), visual stimulus (V), and auditory stimulus (A) from 18 migraine patients and 21 healthy control (HC) groups were used. By applying continuous wavelet transform (CWT) and short-time Fourier transform (STFT) methods to these EEG signals, scalogram-spectrogram images were obtained in the time-frequency (T-F) plane. Then, these images were applied as inputs in three different convolutional neural networks (CNN) architectures (AlexNet, ResNet50, SqueezeNet) that proposed deep convolutional neural network (DCNN) models and classification was performed. The results of the classification process were evaluated, taking into account accuracy (acc.), sensitivity (sens.), specificity (spec.), and performance criteria, and the performances of the preferred methods and models in this study were compared. In this way, the situation, method, and model that showed the most successful performance for the early diagnosis of MD were determined. Although the classification results are close to each other, the resting state, CWT method, and AlexNet classifier showed the most successful performance (Acc: 99.74%, Sens: 99.9%, Spec: 99.52%). We think that the results obtained in this study are promising for the early diagnosis of MD and can be of help to experts.

Comparative study of continuous wavelet transform and multivariate calibration for the simultaneous spectrophotometric determination of tamsulosin and solifenacin in combined dosage form and biological sample.

Spectrophotometry alone is not applicable for the simultaneous determination of drugs in a multi-component pharmaceutical formulation owing to their overlap. In this study, the combination of UV-Vis spectrophotometry and chemometric methods, including continuous wavelet transform (CWT) and partial least squares (PLS) was presented for the simultaneous estimation of tamsulosin (TAM) and solifenacin (SOL) in synthetic mixtures, commercial formulation, and biological sample. The simultaneous spectrophotometric determination of TAM and SOL in binary mixtures, real sample, and biological sample was performed by applying CWT and PLS approaches. In the CWT method, two various wavelet families named Daubechies (db2) at wavelength 223 nm and Biorthogonal (bior1.3) at wavelength 227 nm based on the appropriate zero crossing point were selected for TAM and SOL, respectively. The linear ranges of TAM and SOL were 0.25-4 μg/mL and 10-30 μg/mL, respectively. The limits of detection (LOD) were 0.0459 μg/mL and 0.2085 μg/mL, while the limits of quantitation (LOQ) were 0.3208 μg/mL and 0.6495 μg/mL for TAM and SOL, respectively. The average recovery values of eighteen mixtures were 98.28% and 97.79% for TAM and SOL, respectively. Also, the root mean square error (RMSE) of both components was lower than 2.3. Based on the k-Fold cross-validation in the PLS approach, the optimum number of components related to the TAM and SOL were 9 and 5 with a mean square error prediction (MSEP) of 0.0153 and 0.0370, respectively. The mean recovery values of the test set were found to be 100.09% for TAM and 99.95% for SOL where RMSE values were 0.0064 and 0.0169 for TAM and SOL, respectively. Analysis of variance (ANOVA) was applied to the results of the real sample that there was no significant difference between the proposed methods and the high-performance liquid chromatography (HPLC) as a reference technique. The result obtained revealed that the proposed methods were found to be fast, facile, economical, and precise, which provide a suitable alternative to HPLC technique for the concurrent determination of TAM and SOL in quality control laboratories. UV-Vis spectrophotometry combined with CWT and PLS was developed.Simultaneous analysis of TAM and SOL was performed using the proposed approaches.These methods were implemented on synthetic mixtures, commercial formulation, and biological sample.ANOVA test was used to compare the suggested methods and the HPLC technique.

Tool Wear Condition Monitoring Method Based on Deep Learning with Force Signals.

Tool wear condition monitoring is an important component of mechanical processing automation, and accurately identifying the wear status of tools can improve processing quality and production efficiency. This paper studied a new deep learning model, to identify the wear status of tools. The force signal was transformed into a two-dimensional image using continuous wavelet transform (CWT), short-time Fourier transform (STFT), and Gramian angular summation field (GASF) methods. The generated images were then fed into the proposed convolutional neural network (CNN) model for further analysis. The calculation results show that the accuracy of tool wear state recognition proposed in this paper was above 90%, which was higher than the accuracy of AlexNet, ResNet, and other models. The accuracy of the images generated using the CWT method and identified with the CNN model was the highest, which is attributed to the fact that the CWT method can extract local features of an image and is less affected by noise. Comparing the precision and recall values of the model, it was verified that the image obtained by the CWT method had the highest accuracy in identifying tool wear state. These results demonstrate the potential advantages of using a force signal transformed into a two-dimensional image for tool wear state recognition and of applying CNN models in this area. They also indicate the wide application prospects of this method in industrial production.

Assessment of 2D Digital Image Correlation for Experimental Modal Analysis of Transient Response of Beams Using a Continuous Wavelet Transform Method

The modal parameters of structures, and in particular their mode shapes, are generally determined based on the measurement of accelerometers or laser vibrometers. However, these sensors do not allow the performance of full-field measurements. In this study, the free vibration of a beam triggered by a shock is investigated using a high-speed camera with high image definition. With the help of digital image correlation (DIC), the beam displacement fields are deduced from the images. To analyse the DIC measurement quality, different tests and analyses are performed. First, the systematic errors and uncertainties in the DIC calculation for a simple translation are analysed considering different speckles. Then, tests on two configurations of a vibrating beam are filmed and full-field displacement measurements are computed. The modal parameters of the beam are deduced from these measurements using a continuous wavelet transform method. Particular care was taken to adapt the method to the post-processing of the numerous and noisy signals obtained for these experiments. All the steps of the post-processing are detailed in this paper. Finally, the modal parameters obtained with the proposed method are compared with those obtained in a more classical way using accelerometers and from the beam theory. In particular, the comparison of the signal-to-noise ratio of the different measurement methods is discussed.

Validation of Estimating Stress from Fracture Injection Tests Using Continuous Wavelet Transform with Experimental Data

The article discusses the new technique for fracture closure pressure detection using continuous wavelet transform (CWT). The study focuses on calibrating the CWT technique and comparing it with different techniques for closure detection. According to the article, traditional methods for identifying the closure of hydraulic fracturing operations are based on assumptions that can conflict with one another, resulting in greatly varying approximations of closure pressure and duration. To address this issue, the article employs a set of diagnostic fracture injection tests that utilize the Step-Rate Injection Method for Fracture In-Situ Properties tool (SIMFIP). By directly observing wellbore deformation, the SIMFIP tool determines the minimum principal stress, while strain gauges monitor the opening and closing of fractures during multiple tests. The publicly accessible data are used to evaluate the accuracy of the new closure detection technique using CWT. The findings indicate that the CWT method aligns with measurements of deformation and can identify the impact of intricate closure events and pre-existing natural fractures. In conclusion, the article suggests that the CWT technique shows great potential as an alternative to traditional approaches for detecting closure pressure.

Using Wavelet Analysis and Deep Learning for EMG-Based Hand Movement Signal Classification

In this study; time series electromyography (EMG) data have been classified according to hand movements using wavelet analysis and deep learning. A pre-trained deep CNN (Convolitonal Neural Network-GoogLeNet) has been used in the classification process performed with signal processing, by this way the results can be obtained by continuous wavelet transform and classification methods. The dataset used has been taken from the Machine Learning Repository at the University of California. In the data set; EMG data of 5 healthy individuals, 2 males and 3 females, of the same age (~20-22 years) are available. Data; It consists of grasping spherical objects (Spher), grasping small objects with fingertips (Tip), grasping objects with palms (Palm), grasping thin/flat objects (Lat), grasping cylindrical objects (Cyl) and holding heavy objects (Hook). It is desired to perform 6 hand movements at the same time. While these movements are necessary, speed and power depend on one's will. People perform each movement for 6 seconds and repeat each movement (action) 30 times. The CWT (Continuous Wavelet Transform) method was used to transform the signal into an image. The scalogram image of the signal was created using the CWT method and the generated images were collected in a data set folder. The collected scalogram images have been classified using GoogLeNet, a deep learning network model. With GoogLeNet, results with 97.22% and 88.89% accuracy rates were obtained by classifying the scalogram images of the signals received separately from channel 1 and channel 2 in the data set. The applied model can be used to classify EMG signals in EMG data with high success rate. In this study, 80% of data was used for educational purposes and 20% for validation purposes. In the study, the results of the classification processes have been evaluated separately for first and second channel data.

Emotion Recognition Using Continuous Wavelet Transform and Ensemble of Convolutional Neural Networks through Transfer Learning from Electroencephalogram Signal

Purpose: Emotions are integral brain states that can influence our behavior, decision-making, and functions. Electroencephalogram (EEG) is an appropriate modality for emotion recognition since it has high temporal resolution and is a non-invasive and cheap technique.&#x0D; Materials and Methods: A novel approach based on Ensemble pre-trained Convolutional Neural Networks (ECNNs) is proposed to recognize four emotional classes from EEG channels of individuals watching music video clips. First, scalograms are built from one-dimensional EEG signals by applying the Continuous Wavelet Transform (CWT) method. Then, these images are used to re-train five CNNs: AlexNet, VGG-19, Inception-v1, ResNet-18, and Inception-v3. Then, the majority voting method is applied to make the final decision about emotional classes. The 10-fold cross-validation method is used to evaluate the performance of the proposed method on EEG signals of 32 subjects from the DEAP database.&#x0D; Results:.The experiments showed that applying the proposed ensemble approach in combinations of scalograms of frontal and parietal regions improved results. The best accuracy, sensitivity, precision, and F-score to recognize four emotional states achieved 96.90% ± 0.52, 97.30 ± 0.55, 96.97 ± 0.62, and 96.74 ± 0.56, respectively.&#x0D; Conclusion: So, the newly proposed model from EEG signals improves recognition of the four emotional states in the DEAP database.

Rapid Quantitative Determination Control Male Sexual Dysfunction Drug Content of Sildenafil and Tramadol Mixture by Spectrophotometry Along With Smart Chemometrics Approaches Compared to the HPLC Reference Method.

There are reports for the simultaneous estimation of sildenafil (SIL) and tramadol (TM) using chromatographic techniques. Some limitations of these methods are high cost and time, more usage of pollutant solvents, and complicated instruments. Simultaneous determination of SIL and TM in the anti-sexual dysfunction pharmaceutical formulation using rapid and smart chemometrics approaches along with spectrophotometry. In this study, a spectrophotometric analysis method based on continuous wavelet transform (CWT), partial least-squares (PLS), and radial basis function neural network (RBF-NN) was presented. In the CWT method, the Symlet (Sym2) and Daubechies (db2) wavelet families with wavelengths of 268 and 221 nm were considered the best families for SIL and TM, respectively. The limit of detection (LOD) and limit of quantitation (LOQ) values were found to be 0.0185, 0.0054 μg/mL and 0.0458, 0.3092 μg/mL for SIL and TM, respectively. The mean recovery values were 101.06 and 101.49% for SIL and TM, respectively. The results of PLS revealed that the root-mean-square error (RMSE) of SIL and TM was 0.0286 and 0.0635, respectively. Also, the mean recovery of both drugs was obtained over 99%. The mean square error (MSE) of RBF-NN was 5.88 × 10-29 and 4.61 × 10-27 for SIL and TM, respectively. A statistical comparative study conducted for the proposed methods with the high-performance liquid chromatography (HPLC) technique demonstrated no significant difference between these methods. The suggested approaches were simple, fast, accurate, and affordable. Three smart UV spectrophotometric methods based on chemometrics were presented to determine SIL and TM simultaneously. The efficient chemometrics methods, including CWT, PLS, and RBF-NN, were used. The mentioned methods are rapid, simple, inexpensive, and accurate. The comparison was performed between the proposed models and HPLC technique.

A new automatic bearing fault size diagnosis using time-frequency images of CWT and deep transfer learning methods

Bearings are generally used as bearings or turning elements. Bearings are subjected to high loads and rapid speeds. Furthermore, metal-to-metal contact within the bearing makes it sensitive. In today?s machines, bearing failures disrupt the operation of the system or completely stop the system. Bearing failures that can occur can cause enormous damage to the entire system. Therefore, it is necessary to anticipate bearing failures and to carry out a regular diagnostic examination. Various systems have been developed for fault diagnosis. In recent years, deep transfer learning (DTL) methods are often preferred in current bearing diagnosis models, as they provide time savings and high success rates. Deep transfer learning models also improve diagnosis accuracy under certain conditions by greatly reducing human intervention. Diagnosis at the right time is very important for the sustainability and efficiency of industrial production. A technique based on continuous wavelet transform (CWT) and two dimensional (2D) convolutional neural networks (CNN) is presented in this paper to detect fault size from vibration data of various bearing failure types. Time-frequency (TF) color scalogram images for bearing vibration signals were obtained using the CWT method. Using AlexNet, GoogleNet, Resnet, VGG16, and VGG19 deep transfer learning methods with scalogram images, fault size prediction from vibration signals was performed. Five different transfer deep learning models were used for three different data sets. It was observed that the success rates obtained varied between 96.67% and 100%.

Climate Change Impacts of Sea Level Rise on the Saemangeum Sea Dike

The Saemangeum Sea Dike, completed in 2011 in South Korea, is the longest sea dike in the world. One of the most pressing issues of today, climate change leads to sea-level rise that could increase the risk of sea dike overflow. However, what impact climate change will have on the Saemangeum Sea Dike has not been verified yet. Here, we estimate the impacts of sea-level rise on the Saemangeum Sea Dike, using the Linear Regression Analysis method, the Continuous Wavelet Transform (CWT) method, and the Ensemble Empirical Mode Decomposition (EEMD) method. The sea level data of the Saemangeum area over the last 30 years were analyzed, and the results showed that the maximum rate of sea-level rise of the Saemangeum Sea Dike is at 2.51 mm/year when the EEMD method was applied. It means that the impact of sea-level rise on the Saemangeum sea dike is reasonably low because the sea level rise will exceed the freeboard after 366 years, so there is currently no clear sign that it poses a problem.

A Novel QT-Interval Analysis Method Based on Continuous Wavelet Transform and Philips Algorithm

Abstract QT surveillance is the most vital appliance to detect the possibility of sudden death sourced by using pro-arrhythmic drugs treating abnormal conditions in the heart. The repolarization of ventricles makes QT interval surveillance difficult since noisy conditions and individual cardiac situations. Besides, an automated QT algorithm is crucial due to a manual QT measurement with some disadvantages such as fatigue condition in reading long records. In this study, a fully novel automated method combining Continuous Wavelet Transform and Philips method was established to perform QT interval analysis. Electrocardiogram recordings were obtained from PhyisoNet database marked by manual and standard automated methods. The proposed algorithm had scores of 15.46 and 11.87 millisecond mean error with 11.85 and 9.91 millisecond standard deviation in terms of gold and silver standards, respectively. Also, the entire QT database was utilized in order to test the algorithm performance with the score of 12.89 and 9.76 millisecond mean and standard deviation errors, respectively. The present algorithm performance had scores of −0.21 ± 7.81 at golden standard, and −4.10 ± 18.21 millisecond error for the whole QT database tests, respectively. The proposed algorithm is attained to more stable and robust results with a higher performance than the previous comparable studies.

A deep learning approach in automated detection of schizophrenia using scalogram images of EEG signals

This study presents a method with high accuracy performance that aims to automatically detect schizophrenia (SZ) from electroencephalography (EEG) records. Unlike related literature studies using traditional machine learning algorithms, the features required for the training of the network are automatically extracted from the EEG records in our method. In order to obtain the time frequency features of the EEG signals, the signal was converted into 2D by using the Continuous Wavelet Transform method. This study has the highest accuracy performance in the relevant literature by using 2D time frequency features in automatic detection of SZ disease. It is trained with Visual Geometry Group-16 (VGG16), an advanced convolutional neural networks (CNN) deep learning network architecture, to extract key features found on scalogram images and train the network. The study shows a high success in classifying SZ patients and healthy individuals with a very satisfactory accuracy of 98% and 99.5%, respectively, using two different datasets consisting of individuals from different age groups. Using different techniques [Activization Maximization, Saliency Map, and Gradient-weighted Class Activation Mapping (Grad-CAM)] to visualize the learning outcomes of the CNN network, the relationship of frequency components between SZ and the healthy individual is clearly shown. Moreover, with these interpretable outcomes, the difference between SZ patients and healthy individuals can be distinguished very easily help for expert opinion.