A signal invariant wavelet function selection algorithm

Quantitative Analysis of Mother Wavelet Function Selection for Wearable Sensors-Based Human Activity Recognition

Recent advancements in the Internet of Things (IoT) wearable devices such as wearable inertial sensors have increased the demand for precise human activity recognition (HAR) with minimal computational resources. The wavelet transform, which offers excellent time-frequency localization characteristics, is well suited for HAR recognition systems. Selecting a mother wavelet function in wavelet analysis is critical, as optimal selection improves the recognition performance. The activity time signals data have different periodic patterns that can discriminate activities from each other. Therefore, selecting a mother wavelet function that closely resembles the shape of the recognized activity's sensor (inertial) signals significantly impacts recognition performance. This study uses an optimal mother wavelet selection method that combines wavelet packet transform with the energy-to-Shannon-entropy ratio and two classification algorithms: decision tree (DT) and support vector machines (SVM). We examined six different mother wavelet families with different numbers of vanishing points. Our experiments were performed on eight publicly available ADL datasets: MHEALTH, WISDM Activity Prediction, HARTH, HARsense, DaLiAc, PAMAP2, REALDISP, and HAR70+. The analysis demonstrated in this paper can be used as a guideline for optimal mother wavelet selection for human activity recognition.

THz spectrum processing method based on optimal wavelet selection

Terahertz spectrum is easily interfered by system noise and water-vapor absorption. In order to obtain high quality spectrum and better prediction accuracy in qualitative and quantitative analysis model, different wavelet basis functions and levels of decompositions are employed to perform denoising processing. In this study, the terahertz spectra of wheat samples are denoised using wavelet transform. The compound evaluation indicators (T) are used for systematically analyzing the quality effect of wavelet transform in terahertz spectrum preprocessing. By comparing the optimal denoising effects of different wavelet families, the wavelets of coiflets and symlets are more suitable for terahertz spectrum denoising processing than the wavelets of fejer-korovkin and daubechies, and the performance of symlets 8 wavelet basis function with 4-level decomposition is the optimum. The results show that the proposed method can select the optimal wavelet basis function and decomposition level of wavelet denoising processing in the field of terahertz spectrum analysis.

Identification of heavy metal pollutants in wheat by THz spectroscopy and deep support vector machine

This paper proposes to detect heavy metal pollutants in wheat using terahertz spectroscopy and deep support vector machine (DSVM). Five heavy metal pollutants, arsenic, lead, mercury, chromium, and cadmium, were considered for detection in wheat samples. THz spectral data were pre-processed by wavelet denoising. DSVM was introduced to further enhance the accuracy of the SVM classification model. According to the relationship between the accuracy and the training time with the number of hidden layers ranging from 1 to 4, the model performs the best when the hidden layer network has three layers. Besides, using the back-propagation algorithm to optimize the entire DSVM network. Compared with Deep neural network (DNN) and SVM models, the comprehensive evaluation index of the proposed model optimized by DSVM has the highest accuracy of 91.3 %. It realized the exploration enhanced the classification accuracy of the heavy metal pollutants in wheat.

A Wavelet Packet Transform and Convolutional Neural Network Method Based Ultrasonic Detection Signals Recognition of Concrete

This paper proposes a new intelligent recognition method for concrete ultrasonic detection based on wavelet packet transform and a convolutional neural network (CNN). To validate the proposed data-based method, a case study is presented where the K-fold cross-validation was adopted to produce the performance analysis and classification experiments. Moreover, three evaluation indicators, precision, recall, and F-score, are calculated for analyzing the classification performance of the trained models. As a result, the obtained four-classifying CNN reaches more than 99% detection accuracy while the lowest recognition accuracy is not less than 92.5% on the testing dataset for the six-classifying CNN model. Compared with the existing stochastic configuration network (SCN) models, the presented method achieves the design objective with better recognition performance. The calculation results of the six-classifying and five-classifying models and related research clearly indicate the remaining challenging tasks for intelligent recognition algorithms in extracting features and classifying mass data from various concrete defects precisely and efficiently.

Automated discretization of 'transpiration restriction to increasing VPD' features from outdoors high-throughput phenotyping data

BACKGROUND

Restricting transpiration under high vapor pressure deficit (VPD) is a promising water-saving trait for drought adaptation. However, it is often measured under controlled conditions and at very low throughput, unsuitable for breeding. A few high-throughput phenotyping (HTP) studies exist, and have considered only maximum transpiration rate in analyzing genotypic differences in this trait. Further, no study has precisely identified the VPD breakpoints where genotypes restrict transpiration under natural conditions. Therefore, outdoors HTP data (15 min frequency) of a chickpea population were used to automate the generation of smooth transpiration profiles, extract informative features of the transpiration response to VPD for optimal genotypic discretization, identify VPD breakpoints, and compare genotypes.

RESULTS

Fifteen biologically relevant features were extracted from the transpiration rate profiles derived from load cells data. Genotypes were clustered (C1, C2, C3) and 6 most important features (with heritability > 0.5) were selected using unsupervised Random Forest. All the wild relatives were found in C1, while C2 and C3 mostly comprised high TE and low TE lines, respectively. Assessment of the distinct p-value groups within each selected feature revealed highest genotypic variation for the feature representing transpiration response to high VPD condition. Sensitivity analysis on a multi-output neural network model (with R of 0.931, 0.944, 0.953 for C1, C2, C3, respectively) found C1 with the highest water saving ability, that restricted transpiration at relatively low VPD levels, 56% (i.e. 3.52 kPa) or 62% (i.e. 3.90 kPa), depending whether the influence of other environmental variables was minimum or maximum. Also, VPD appeared to have the most striking influence on the transpiration response independently of other environment variable, whereas light, temperature, and relative humidity alone had little/no effect.

CONCLUSION

Through this study, we present a novel approach to identifying genotypes with drought-tolerance potential, which overcomes the challenges in HTP of the water-saving trait. The six selected features served as proxy phenotypes for reliable genotypic discretization. The wild chickpeas were found to limit water-loss faster than the water-profligate cultivated ones. Such an analytic approach can be directly used for prescriptive breeding applications, applied to other traits, and help expedite maximized information extraction from HTP data.