Peak Detection Algorithm Research Articles

Combination of continuous wavelet transform and genetic algorithm-based Otsu for efficient mass spectrometry peak detection

Mass spectrometry (MS) data is susceptible to random noises and alternating baseline, posing great challenges to spectral peak detection, especially for weak peaks and overlapping peaks. Herein, an efficient peak detection algorithm combining continuous wavelet transform (CWT) and genetic algorithm-based threshold segmentation (denoted as WSTGA) for mass spectrometry was proposed. Firstly, Mexican Hat wavelet was selected as the mother wavelet by comparing the matching degree between the difference of Gaussian (DOG) and different wavelets. Subsequently, the ridges and valleys were identified from 2D wavelet coefficient matrix. Afterward, an improved threshold segmentation method, Otsu method based on genetic algorithm, was introduced to find optimal segmentation threshold and achieve better image segmentation, overcoming the deficiency of traditional Otsu method that cannot handle long-tailed unimodal histograms. Finally, the characteristic peaks were successfully identified by utilizing the ridge-valley lines in wavelet space and original spectrum. Receiver operating characteristic (ROC) curve, area under curve (AUC) and F₁ measure are used as criterions to evaluate performance of peak detection algorithms. Compared with multi-scale peak detection (MSPD) and CWT and image segmentation (CWT-IS) methods, all the results showed that WSTGA can achieve better peak detection. More importantly, the experimental results from MALDI-TOF spectra demonstrated that WSTGA can effectively detect more weak peaks and overlapping peaks while maintaining a lower false peak detection rate than MSPD and CWT-IS methods, indicating its great advantages in characteristic peak identification.

An optimized algorithm for peak detection in noisy periodic and quasi-periodic signals

This paper represents a modified version (called T-AMPD) of automatic multiscale-based peak detection (AMPD) algorithm, which was previously proposed by Scholkmann et al for peak detection of noisy periodic and quasi-periodic signals, in order to optimize the algorithm and reduce the computational cost while still maintaining the accuracy. By introducing a new parameter , the number of expected peaks, to the algorithm and some other adjustments, the T-AMPD outperforms the original algorithm in processing time. Experiments in some real-world signals show that the modified algorithm is greatly better than the original one in runtime performance. Specifically, our T-AMPD was compared to the newest version of AMPD which is available in an R programming package. As a result, our T-AMPD is 2 to 25 times faster than the original AMPD in computational time, while the accuracy remains unchanged.

Estimation of Step Length With Wearable Thigh Sensor Using an Unscented Kalman Filter.

The determination of step length, an important gait parameter, has been a challenging task. Although unobtrusive sensors (inertial measurement units) have been developed recently, they cannot facilitate the automatic estimation of step length. In this article, we use a model-based technique to determine the step length using the Unscented Kalman Filter with angular velocity from a gyroscope inside the thigh pocket. We then propose a novel covariance estimation algorithm based on a screening technique that performs a search for the optimal Process Noise Covariance matrix. Upon implementing the Unscented Kalman Filter, the step length is found using the horizontal position of the foot relative to the hip using a patient-independent robust peak detection algorithm. This research article paves the way for algorithms that are computationally much faster than black box methods, with more scope for the development of better algorithms for covariance estimation using the one proposed in this article as a foundation.

A universal, high-performance ECG signal processing engine to reduce clinical burden.

BackgroundElectrocardiogram (ECG) signal conditioning is a vital step in the ECG signal processing chain that ensures effective noise removal and accurate feature extraction.ObjectiveThis study evaluates the performance of the FDA 510 (k) cleared HeartKey Signal Conditioning and QRS peak detection algorithms on a range of annotated public and proprietary ECG databases (HeartKey is a UK Registered Trademark of B‐Secur Ltd).MethodsSeven hundred fifty‐one raw ECG files from a broad range of use cases were individually passed through the HeartKey signal processing engine. The algorithms include several advanced filtering steps to enable significant noise removal and accurate identification of the QRS complex. QRS detection statistics were generated against the annotated ECG files.ResultsHeartKey displayed robust performance across 14 ECG databases (seven public, seven proprietary), covering a range of healthy and unhealthy patient data, wet and dry electrode types, various lead configurations, hardware sources, and stationary/ambulatory recordings from clinical and non‐clinical settings. Over the NSR, MIT‐BIH, AHA, and MIT‐AF public databases, average QRS Se and PPV values of 98.90% and 99.08% were achieved. Adaptable performance (Se 93.26%, PPV 90.53%) was similarly observed on the challenging NST database. Crucially, HeartKey's performance effectively translated to the dry electrode space, with an average QRS Se of 99.22% and PPV of 99.00% observed over eight dry electrode databases representing various use cases, including two challenging motion‐based collection protocols.ConclusionHeartKey demonstrated robust signal conditioning and QRS detection performance across the broad range of tested ECG signals. It should be emphasized that in no way have the algorithms been altered or trained to optimize performance on a given database, meaning that HeartKey is potentially a universal solution capable of maintaining a high level of performance across a broad range of clinical and everyday use cases.

Batch alignment via retention orders for preprocessing large-scale multi-batch LC-MS experiments.

Meticulous selection of chromatographic peak detection parameters and algorithms is a crucial step in preprocessing liquid chromatography-mass spectrometry (LC-MS) data. However, as mass-to-charge ratio and retention time shifts are larger between batches than within batches, finding apt parameters for all samples of a large-scale multi-batch experiment with the aim of minimizing information loss becomes a challenging task. Preprocessing independent batches individually can curtail said problems but requires a method for aligning and combining them for further downstream analysis. We present two methods for aligning and combining individually preprocessed batches in multi-batch LC-MS experiments. Our developed methods were tested on six sets of simulated and six sets of real datasets. Furthermore, by estimating the probabilities of peak insertion, deletion and swap between batches in authentic datasets, we demonstrate that retention order swaps are not rare in untargeted LC-MS data. kmersAlignment and rtcorrectedAlignment algorithms are made available as an R package with raw data at https://metabocombiner.img.cas.cz. Supplementary data are available at Bioinformatics online.

SpO2 Measurement: Non-Idealities and Ways to Improve Estimation Accuracy in Wearable Pulse Oximeters

The blood oxygen saturation level (SpO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> ) has become one of the vital body parameters for the early detection, monitoring, and tracking of the symptoms of coronavirus diseases 2019 (COVID-19) and is clinically accepted for patient care and diagnostics. Pulse oximetry provides non-invasive SpO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> monitoring at home and ICUs without the need of a physician/doctor. However, the accuracy of SpO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> estimation in wearable pulse oximeters remains a challenge due to various non-idealities. We propose a method to improve the estimation accuracy by denoising the red and IR signals, detecting the signal quality, and providing feedback to hardware to adjust the signal chain parameters like LED current or transimpedance amplifier gain and enhance the signal quality. SpO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> is calculated using the red and infrared photoplethysmogram (PPG) signals acquired from the wrist using Texas Instruments AFE4950EVM. We introduce the green PPG signal as a reference to obtain the window size of the moving average filter for baseline wander removal and as a timing reference for peak and valley detection in the red and infrared PPG signals. We propose the improved peak and valley detection algorithm based on the incremental merge segmentation algorithm. Kurtosis, entropy, and Signal-to-noise ratio (SNR) are used as signal quality parameters, and SNR is further related to the variance in the SpO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> measurement. A closed-loop implementation is performed to enhance signal quality based on the signal quality parameters of the recorded PPG signals. The proposed algorithm aims to estimate SpO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> with a variance of 1% for the pulse oximetry devices.

Flexible and fast estimation of binary merger population distributions with an adaptive kernel density estimator

The LIGO Scientific, Virgo and KAGRA Collaborations recently released the third gravitational wave transient catalog or GWTC-3, significantly expanding the number of gravitational wave signals. To address the---still uncertain---formation channels of the source compact binaries, their population properties must be characterized. The computational cost of the Bayesian hierarchical methods employed thus far scales with the size of the event catalogs, and such methods have until recently assumed fixed functional forms for the source distribution. Here we propose a fast and flexible method to reconstruct the population of LIGO-Virgo merging black hole (BH) binaries without such assumptions. For sufficiently high event statistics and sufficiently low individual event measurement error (relative to the scale of population features) a kernel density estimator (KDE) reconstruction of the event distribution will be accurate. We improve the accuracy and flexibility of KDE for finite event statistics using an adaptive bandwidth KDE (awKDE). We apply awKDE to publicly released parameter estimates for 44 significant (69) BH binary mergers in GWTC-2 (GWTC-3), in combination with a fast polynomial fit of search sensitivity, to obtain a nonparametric estimate of the mass distribution, and compare to Bayesian hierarchical methods. We also demonstrate a robust peak detection algorithm based on awKDE and use it to calculate the significance of the apparent peak in the BH mass distribution around $35\text{ }\text{ }{\mathrm{M}}_{\ensuremath{\bigodot}}$. We find such a peak is very unlikely to have occurred if the true distribution is a featureless power-law (significance of $3.6\ensuremath{\sigma}$ for confident GWTC-2 BBH events, $3.0\ensuremath{\sigma}$ for confident GWTC-3 BBH events).

Minimizing the coincidence error in particle size spectrometers with digital signal processing techniques

Aerosol quantification is highly coveted for many applications ranging from quality monitoring to healthcare. Optical particle size spectrometers are one of the popular measurement devices used in aerosol characterization, which provide information about the size distribution and concentration of the particles. One drawback of this method is that the spectrometers are limited to applications with relatively low concentrations. At high concentrations, the number concentration is underestimated, while the size distribution is shifted towards larger particles. This phenomenon is commonly referred as coincidence error. However, the counting efficiency, above the lower detection limit, depends i. a. on the detector dead time. In order to minimize the probability of coincidence events, the dead time of the detector has to be as small as possible. The contribution of this work is to minimize the coincidence error by purposefully reducing the detector dead time. Therefore, various digital signal processing methods, to minimize dead time, are investigated and experimentally verified. Wavelet and derivative-based peak detection algorithms were found to be able to separate even strongly overlapping scattered light pulses from each other. The results show that the detector dead time can be reduced by 65% with digital signal processing techniques. Allowing the quantification of aerosols with more than 2.9 times the maximum concentration of current optical aerosol spectrometers, without increasing the coincidence error.

Automated visual tracking of crack growth in coupon and component level fatigue testing using thermoelastic stress analysis

An automated system capable of visually detecting the two dimensional location of a crack in metal is presented. A peak-detection algorithm is applied to the thermoelastic signal component caused by heat diffusion, obtained from thermoelastic stress analysis using an industrial microbolometer, which is mounted on motion stages, allowing two dimensional tracking of the crack. The system is experimentally validated on coupons under constant amplitude loading, with the crack tip location estimates compared against measurements from an optical microscope. The system is also applied to monitoring crack growth on a geometrically complex section of an aircraft wing under representative flight-spectrum loading.

Application of a hand-held laser methane detector for measuring enteric methane emissions from cattle in intensive farming.

The hand-held laser methane detector (LMD) technique has been suggested as an alternative method for measuring methane (CH4) emissions from enteric fermentation of ruminants in the field. This study aimed to establish a standard procedure for using LMD to assess CH4 production in cattle and evaluate the efficacy of the protocol to detect differences in CH4 emissions from cattle fed with diets of different forage-to-concentrate (FC) ratios. Experiment 1 was conducted with four Hanwoo steers (584 ± 57.4 kg body weight [BW]) individually housed in metabolic cages. The LMD was installed on a tripod aimed at the animal's nostril, and the CH4 concentration in the exhaled gas was measured for 6 min every hour for 2 consecutive days. For the data processing, the CH4 concentration peaks were identified by the automatic multi-scale peak detection algorithm. The peaks were then separated into those from respiration and eructation by fitting combinations of two of the four distribution functions (normal, log-normal, gamma, and Weibull) using the mixdist R package. In addition, the most appropriate time and number of consecutive measurements to represent the daily average CH4 concentration were determined. In experiment 2, 30 Hanwoo growing steers (343 ± 24.6 kg BW), blocked by BW, were randomly divided into three groups. Three different diets were provided to each group: high FC ratio (35:65) with low-energy concentrate (HFC-LEC), high FC ratio with high-energy concentrate (HFC-HEC), and low FC ratio (25:75) with high-energy concentrate (LFC-HEC). After 10 d of feeding the diets, the CH4 concentrations for all steers were measured and analyzed in duplicate according to the protocol established in experiment 1. In experiment 1, the mean correlation coefficient between the CH4 concentration from respiration and eructation was highest when a combination of two normal distributions was assumed (r = 0.79). The most appropriate measurement times were as follows: 2 h and 1 h before, and 1 h and 2 h after morning feeding. Compared with LFC-HEC, HFC-LEC showed 49% and 57% higher CH4 concentrations in exhaled gas from respiration and eructation (P < 0.01). In conclusion, the LMD method can be applied to evaluate differences in CH4 emissions in cattle using the protocol established in this study.

A hybrid genetic algorithm and grey wolf optimizer technique for faster global peak detection in PV system under partial shading

In the present scenario, renewable energy is the fastest growing trend in the energy sector. Chief among the renewable power sources is solar photovoltaic (PV) technology. One of the major problems of solar PV systems is the occurrence of a partial shadow over the PV array, which generates multiple peaks in the power versus voltage (P-V) characteristics of the PV array. Under these conditions, conventional Maximum Power Point Tracking (MPPT) techniques tend to get stuck on local peaks and fail to track the Global Maximum Power Point (GMPP). In this paper, an improved hybrid optimization algorithm named as Genetic Algorithm-Grey Wolf Optimizer (GA-GWO) algorithm, which combines the abilities of the Genetic Algorithm (GA) and Grey Wolf Optimizer (GWO) algorithms, has been proposed to track the global peak. The proposed system has been modeled and examined using MATLAB/Simulink software for various partial and uniform shading conditions. It has also been compared with previously existing GWO and GA algorithms. The proposed algorithm is found to offer faster and better tracking of GMPP than both GWO and GA algorithms.

Research on Coded Excitation Using Kasami Sequence in the Long Rail Detection Based on UGW

For a broken rail detection system based on ultrasonic guided waves (UGW), the multimodal and dispersion characteristics of UGW degrade signal-to-noise ratio (SNR) and range resolution. To improve the SNR of the received signals and range resolution, the coded excitation based on Kasami sequences is presented in this work. Utilizing a PSpice model of piezoelectric ultrasonic transducers, as well as conducting field tests based on the pitch–catch mechanism, it is shown that encoded UGW signals can increase the SNRG (the gain of SNR) by 6.29 dB. The main lobe width of the coded excitation is mainly determined by the number of carrier cycles and the carrier waveform, and the size of the side lobes is mainly determined by the number of coding bits. To quickly identify the corresponding transmissions at the receivers, a peak detection algorithm is shown. It is based on bandpass filter, triangle filter and Hilbert transform. Its accuracy and effectiveness are validated by using some field tests under different distances. It can be concluded that the shown adaptive peak algorithm has strong robustness and immunity to noise.

Explainable detection of myocardial infarction using deep learning models with Grad-CAM technique on ECG signals

Myocardial infarction (MI) accounts for a high number of deaths globally. In acute MI, accurate electrocardiography (ECG) is important for timely diagnosis and intervention in the emergency setting. Machine learning is increasingly being explored for automated computer-aided ECG diagnosis of cardiovascular diseases. In this study, we have developed DenseNet and CNN models for the classification of healthy subjects and patients with ten classes of MI based on the location of myocardial involvement. ECG signals from the Physikalisch-Technische Bundesanstalt database were pre-processed, and the ECG beats were extracted using an R peak detection algorithm. The beats were then fed to the two models separately. While both models attained high classification accuracies (more than 95%), DenseNet is the preferred model for the classification task due to its low computational complexity and higher classification accuracy than the CNN model due to feature reusability. An enhanced class activation mapping (CAM) technique called Grad-CAM was subsequently applied to the outputs of both models to enable visualization of the specific ECG leads and portions of ECG waves that were most influential for the predictive decisions made by the models for the 11 classes. It was observed that Lead V4 was the most activated lead in both the DenseNet and CNN models. Furthermore, this study has also established the different leads and parts of the signal that get activated for each class. This is the first study to report features that influenced the classification decisions of deep models for multiclass classification of MI and healthy ECGs. Hence this study is crucial and contributes significantly to the medical field as with some level of visible explainability of the inner workings of the models, the developed DenseNet and CNN models may garner needed clinical acceptance and have the potential to be implemented for ECG triage of MI diagnosis in hospitals and remote out-of-hospital settings.

A ‘shape-orientated’ algorithm employing an adapted Marr wavelet and shape matching index improves the performance of continuous wavelet transform for chromatographic peak detection and quantification

A new ‘shape-orientated’ continuous wavelet transform (CWT)-based algorithm employing an adapted Marr wavelet (AMW) with a shape matching index (SMI), defined as peak height normalized wavelet coefficient (C′max/H) for feature filtering, was developed for chromatographic peak detection and quantification. Exploiting the chromatographic profile of a candidate peak, AMW-SMI algorithm emphasizes more on the matching of the overall chromatographic profile to a reference Gaussian shape, while partly alleviates the requirement on the signal intensity derived from single or several data points, thus it allows the detection of low-intensity features from metabolites at low abundance. AMW-SMI imposes maximum and minimum thresholds on the ridgeline width and length to define a valid ridgeline, which corresponds to a more stably shaped chromatographic profile. The maximum wavelet coefficient Cmax′(a0,b0) on the valid ridgeline determines the translation b0 as the peak center. AMW-SMI detects the valley lines to define the peak boundaries, which is important to obtain accurate peak quantification. As a more ‘shape-orientated’ peak detection algorithm, various methods related to the ‘shape’ are introduced for feature filtering, out of which, the effective SNR (eSNR) is defined to evaluate if the shape is strong or good enough relative to the ‘shape noise’, and the SMI, which can quantitatively evaluate the shape quality regardless of the data intensities and peak width, is applied to filter out the poorly shaped false positives. AMW-SMI performs Gaussian fitting of all data points between the defined peak boundaries to refine the peak parameters, and the refined SMI, SNR and peak width can be applied for further feature filtering and reinforce the ‘shape-quality’ of final selected peaks. The performance of AMW-SMI is evaluated qualitatively (by recall, precision and F-score) and quantitatively (by ratio of isotopic features and triplicate RSD) on the LC-MS data of model mixtures of 21 human metabolite standards and 8 plant metabolite standards, and of serum sample spiked with the 21 human metabolite standards, and on the triplicate LC-MS data of the same sample of cell metabolomic extracts. Compared with XCMS (centWave) and MZmine 2 (ADAP), the proposed AMW-SMI algorithm can faithfully identify chromatographic peaks with significantly fewer false positives and demonstrated general superiority in terms of qualitative precision (robustness) and quantitative accuracy (by ratio of isotopic features), and comparable recall (sensitivity) and quantitative stability (by RSD of triplicate measurements).

Convolutional neural network for automated peak detection in reversed-phase liquid chromatography

Although commercially available software provides options for automatic peak detection, visual inspection and manual corrections are often needed. Peak detection algorithms commonly employed require carefully written rules and thresholds to increase true positive rates and decrease false positive rates. In this study, a deep learning model, specifically, a convolutional neural network (CNN), was implemented to perform automatic peak detection in reversed-phase liquid chromatography (RPLC). The model inputs a whole chromatogram and outputs predicted locations, probabilities, and areas of the peaks. The obtained results on a simulated validation set demonstrated that the model performed well (ROC-AUC of 0.996), and comparably or better than a derivative-based approach using the Savitzky-Golay algorithm for detecting peaks on experimental chromatograms (8.6% increase in true positives). In addition, predicted peak probabilities (typically between 0.5 and 1.0 for true positives) gave an indication of how confident the CNN model was in the peaks detected. The CNN model was trained entirely on simulated chromatograms (a training set of 1,000,000 chromatograms), and thus no effort had to be put into collecting and labeling chromatograms. A potential major drawback of this approach, namely training a CNN model on simulated chromatograms, is the risk of not capturing the actual “chromatogram space” well enough that is needed to perform accurate peak detection in real chromatograms.

Deep Learning Approach for Fixed and Rotary-Wing Target Detection and Classification in Radars

Recent advancements demonstrate the application of machine learning techniques to radar signal processing, particularly the use of neural networks in one-dimensional (1-D) radar data to assist constant false alarm rate (CFAR) based target detection. However, the efficacy of using neural networks alone on 2-D data for target extraction remains to be explored. This research aims to develop a novel approach of designing a convolution neural network (CNN) based target detection system for real radar data. We propose a dual neural network detection scheme in which the first neural net filters out noisy images. Thereafter, we employ a second CNN-based target detector that extracts information from both radar data axes (fast-time and slow-time), thus utilizing additional information for better detection of targets even with low SNR. Moreover, the same CNN, in conjunction with a region-based convolutional network, exploits micro-Doppler features of rotary-wing aircraft in radar returns and uses them to further classify detected targets as fixed-wing or rotary-wing aircraft. To the best of our knowledge, no generalized methodology exists to address these two issues together. The proposed solution replaces the critical CFAR and peak detection algorithms from the radar signal processing chain. This article also introduces a labeled range-Doppler images dataset obtained from real-world air traffic control radar data. Experimental results demonstrate the superior performance of the proposed technique over 1-D and 2-D CFAR and peak detection algorithms.

Probabilistic spatial analysis in quantitative microscopy with uncertainty-aware cell detection using deep Bayesian regression.

The investigation of biological systems with three-dimensional microscopy demands automatic cell identification methods that not only are accurate but also can imply the uncertainty in their predictions. The use of deep learning to regress density maps is a popular successful approach for extracting cell coordinates from local peaks in a postprocessing step, which then, however, hinders any meaningful probabilistic output. We propose a framework that can operate on large microscopy images and output probabilistic predictions (i) by integrating deep Bayesian learning for the regression of uncertainty-aware density maps, where peak detection algorithms generate cell proposals, and (ii) by learning a mapping from prediction proposals to a probabilistic space that accurately represents the chances of a successful prediction. Using these calibrated predictions, we propose a probabilistic spatial analysis with Monte Carlo sampling. We demonstrate this in a bone marrow dataset, where our proposed methods reveal spatial patterns that are otherwise undetectable.

A Real-Time PPG Peak Detection Method for Accurate Determination of Heart Rate during Sinus Rhythm and Cardiac Arrhythmia.

Objective: We have developed a peak detection algorithm for accurate determination of heart rate, using photoplethysmographic (PPG) signals from a smartwatch, even in the presence of various cardiac rhythms, including normal sinus rhythm (NSR), premature atrial contraction (PAC), premature ventricle contraction (PVC), and atrial fibrillation (AF). Given the clinical need for accurate heart rate estimation in patients with AF, we developed a novel approach that reduces heart rate estimation errors when compared to peak detection algorithms designed for NSR. Methods: Our peak detection method is composed of a sequential series of algorithms that are combined to discriminate the various arrhythmias described above. Moreover, a novel Poincaré plot scheme is used to discriminate between basal heart rate AF and rapid ventricular response (RVR) AF, and to differentiate PAC/PVC from NSR and AF. Training of the algorithm was performed only with Samsung Simband smartwatch data, whereas independent testing data which had more samples than did the training data were obtained from Samsung’s Gear S3 and Galaxy Watch 3. Results: The new PPG peak detection algorithm provides significantly lower average heart rate and interbeat interval beat-to-beat estimation errors—30% and 66% lower—and mean heart rate and mean interbeat interval estimation errors—60% and 77% lower—when compared to the best of the seven other traditional peak detection algorithms that are known to be accurate for NSR. Our new PPG peak detection algorithm was the overall best performers for other arrhythmias. Conclusion: The proposed method for PPG peak detection automatically detects and discriminates between various arrhythmias among different waveforms of PPG data, delivers significantly lower heart rate estimation errors for participants with AF, and reduces the number of false negative peaks. Significance: By enabling accurate determination of heart rate despite the presence of AF with rapid ventricular response or PAC/PVCs, we enable clinicians to make more accurate recommendations for heart rate control from PPG data.

Pink noise removal and spectral distortion correction based fiber Bragg grating demodulation algorithm.

Pink noise and spectral distortions affect the demodulation accuracies of fiber Bragg gratings (FBGs). A noise reduction algorithm with Savitzky-Golay filtering on each intrinsic mode function component decomposed by CEEMDAN, followed by signal reconstruction from the filtered components, is presented herein. Simulations show that the algorithm retains FBG signal details while reducing noise, avoids measurement errors from excessive smoothing, and suppresses white and pink noises adequately. A resolution-enhanced peak detection algorithm with distortion spectrum correction is also proposed with high accuracy and simple calculation. The experimental results show that the goodness of fit of the FBG temperature-wavelength curve R2 improves from 0.9826 to 0.9999.

An Improved Algorithm for Peak Detection Based on Weighted Continuous Wavelet Transform

Peak detection is a crucial preprocessing step in the analysis of various spectral signals. The method based on the continuous wavelet transform is more practical and popular, and has better detection accuracy and reliability by identifying peaks across scales in the wavelet space and implicitly removing noise as well as the baseline. However, there are inevitably overlapping peaks in the measured spectra, and the formed composite ridges affect peak detection accuracy. Most peak detection methods have limited applicability to overlapping peaks. A weighted continuous wavelet transform (WCWT) peak detection algorithm is proposed to improve the adaptive ability of the peak detection method. This method yields more obvious spectral peak characteristics in low-scale regions. Composite ridges can be successfully truncated by setting a noise threshold based on the standard deviation of the spectral signal. In addition, the maximum value in the ridges was compressed and shifted to a smaller scale, which could determine the peaks more accurately. The method was applied to the peak detection of simulated spectra, Romanian database of Raman spectra, and real liquid electrode glow discharge spectra. The results show that the proposed method exhibits good peak detection performance.