Wavelet Filtering Technique Research Articles

Application of Machine Learning Algorithm for Defect Analysis in Semiconductors Using High Resolved Scanning Acoustic Microscopy

Scanning Acoustic Microscopy is a dedicated powerful technology in NDT to analyze and characterize bonded interfaces for different technologies. As the complexity of the semiconductor package grows, even seemingly insignificant faults become quite important. Main applications include detection of voids, inclusions and delaminated areas in bonding interfaces and thickness variations in layer structures. The SAM operates with the pulse reflection method. The acoustic lens transforms high frequency electromagnetic vibrations that are transmitted inside the lens as a planar parallel wave field. Through the coupling medium, which is water, the cavity concentrates the sound field on the sample. The sound pulses reflected from the sample are gathered up by the acoustic lens and are converted into electromagnetic pulses by the transducer. These pulses are reconstructed into images with defined gray values. Pixels with certain gray levels are used to depict pulses.New developments in SAM combine high throughput inspection and high resolution down to the sub-micrometer range.The wafer bow is quickly eliminated by altering the array transducer focus point throughout the entire scanning procedure. New techniques for SAM's bonded interface protection keep the wafer almost completely dry throughout inspection. In order to discover and categorize failures with the least amount of data, research focuses on deep learning networks and scanning acoustic microscopes for automated failure assessment. High speed automated wafer-level inspections were made possible by the development of image and signal-based machine learning techniques. These operations can be carried out at both the wafer level and the die level without the need for sample preparation.The automatic handling, scanning, drying, and separation of good wafers from defected wafers based on acoustic data made possible by technological developments in SAM systems. Despite the high sensitivity, minute variations in the reflected signal that would point to a flaw can be lost during image reconstruction. More emphasis is being placed on creating AI-based algorithms for highly accurate automated signal interpretation for failure detection as a result of recent breakthroughs in AI (Artificial Intelligence). Since it requires less human work, automated failure analysis is becoming more and more popular for the detection and qualification of wafer bond integrity. This new technique trains the network for automated fault prediction using raw A-Scan data. In the first phase, die level prediction is reviewed and in the later phase, a wafer-level high volume inspection is performed. Combining SAM with an A-Scan-based Deep Learning approach primarily benefits non-destructive analysis with less labelling effort and training time. With minimal human effort, this process flow is perfect for automated high-volume inspection.SAM also uses image based machine learning approach for automated defect analysis. The U-net machine-learning algorithm serves as the foundation for the integrated image analysis program. Deep Learning technologies is integrated into a new modular image analysis (MIA) architecture with built-up integrated pipeline for Label Studio Software-based Deep Learning activities (labelling, training, analysis). The built-up pipeline is used to identify, train, and test various classes of defect in a bonded wafer. This method employs model confidence to manually mark marginal and defective predictions.High resolution and high throughput analysis for the defect characterization and defect analysis of various types of bonded wafers is improved in scanning acoustic microscopes using the array transducers, signal processing methods like SAFT, Wavelet filtering techniques, Artificial intelligence methods based on both image and signal. With all these advanced technologies, Scanning Acoustic Microscopy is capable of inspection of wafer bond integrity, TSV characterization, 3D package evaluation, Wafer Level Chip Scale Packaging (WL-CSP) and Crack inspection. For the metrology of mass production, from initial research and development, our equipment services and software algorithms assist manufacturers of nano electronics in better defect analysis and material characterization. Next Generation transducers were designed and developed in particular for WL-CSP applications.

Prediction of Lost Circulation in Southwest Chinese Oil Fields Applying Improved WOA-BiLSTM

Drilling hazards can be significantly decreased by anticipating potential mud loss and then putting the right well control measures in place. Therefore, it is critical to provide early estimates of mud loss. To solve this problem, an enhanced WOA (Whale Optimization Algorithm) and a BiLSTM (Bidirectional Long Short Term Memory) optimization based prediction model of lost circulation prior to drilling has been created. In order to minimize the noise in the historical comprehensive logging data, a wavelet filtering technique was first used. Then, according to the nonlinear Spearman rank correlation coefficient between mud loss and logging parameter values from large to small, seven characteristic parameters were preferred, and the sliding window was used to extract the relevant data. Secondly, the number of neurons in the first and second hidden layers, the maximum training time, and the initial learning rate of the BiLSTM model were optimized using the enhanced WOA method. The BiLSTM network was given the acquired superparameters in order to improve the model’s ability to predict occurrences. Finally, the model was trained and tested using the processed data. In comparison to the LSTM model, BiLSTM model, and WOA-BiLSTM model, respectively, the improved WOA-BiLSTM early mud loss prediction in southwest Chinese oil fields suggested in this study beat the others, receiving 22.3%, 18.7%, and 4.9% higher prediction accuracy, respectively.

Multi-source information fusion-based dynamic model for online prediction of rate of penetration (ROP) in drilling process

Rate of penetration (ROP) accurate prediction can effectively reduce drilling cycle time and operation cost. However, the complexity and variability of drilling process is not conducive to accurate ROP modeling. The existing ROP models focus on static prediction considering drilling data, which cannot effectively solve the ROP mutation problem caused by complex formations. Some ROP online prediction studies have great real-time capture capability, but without considering the formation lithology adequately, and can be limited in practical application. In this paper, a multi-source information fusion-based dynamic model for online prediction of ROP in drilling process is proposed, which consists of three stages (drilling data pre-processing, optimized ROP modeling, and ROP model updating). In the first stage, three drilling parameters including weight on bit, rotational speed, and torque are filtered by using the outlier removing and wavelet filtering techniques. In the second stage, a novel ROP modeling method, named as hybrid bat algorithm optimized - restricted Boltzmann machine - back propagation neural network (HBAO-RBM-BPNN) is proposed to build the ROP prediction model. In the last stage, formation drillability variation and time interval are used as update conditions, and the established ROP model is updated via a moving window strategy to realize high accuracy online prediction of ROP. Multi-source (formation and drilling) information fusion is considered in this key stage. Finally, the proposed method and seven ROP prediction methods (two offline and five online) are compared through the use of industrial data from a drilling site in Dandong area, Northeast China. The comparison results verify the effectiveness of the proposed method, which lays a foundation for intelligent optimization control of complex geological drilling process.

Deep Learning-Empowered Digital Twin Using Acoustic Signal for Welding Quality Inspection

Weld site inspection is a research area of interest in the manufacturing industry. In this study, a digital twin system for welding robots to examine various weld flaws that might happen during welding using the acoustics of the weld site is presented. Additionally, a wavelet filtering technique is implemented to remove the acoustic signal originating from machine noise. Then, an SeCNN-LSTM model is applied to recognize and categorize weld acoustic signals according to the traits of strong acoustic signal time sequences. The model verification accuracy was found to be 91%. In addition, using numerous indicators, the model was compared with seven other models, namely, CNN-SVM, CNN-LSTM, CNN-GRU, BiLSTM, GRU, CNN-BiLSTM, and LSTM. A deep learning model, and acoustic signal filtering and preprocessing techniques are integrated into the proposed digital twin system. The goal of this work was to propose a systematic on-site weld flaw detection approach encompassing data processing, system modeling, and identification methods. In addition, our proposed method could serve as a resource for pertinent research.

Wheel Bearing Fault Detection for Automobiles using Wheel Speed Sensor

Wheel bearing fault diagnosis and monitoring for automobiles are getting more attention as the market shifts towards electric vehicles (EV) and autonomous vehicles (AV). Electric drivetrains are expected to have longer lifespan than the reliable life of low-cost wheel bearing designs. The probability of having a failed wheel bearing in an EV thus increases. In an AV setting, most of the rides are taken by people who are unfamiliar to the vehicle, removing the human as a viable sensor to detect wheel bearing defects. Fleet managers will be required to pay a high price to inspect potential defects. The implementation of a wheel bearing fault detection system can help reduce these costs.
 In our previous work, a method for injecting wheel bearing brinelling failure was developed. In this paper, we propose a wheel bearing brinelling fault detection algorithm using wheel speed sensor. The proposed algorithm detects wheel bearing faults by identifying peaks in the wheel speed spectrum at bearing critical frequencies. To implement this approach, we leverage phase domain transform technique to normalize against effect of varying speed. A wavelet filtering technique is employed to enhance the bearing fault signatures prior transforming the wheel speed signal into frequency domain. Post-processing techniques are developed to smooth the high-variance wheel speed spectra and improve the detectability of the wheel bearing faults. A regression model is then developed to calculate the wheel bearing health indicators. The results show that the proposed algorithm achieves high performance in detecting wheel bearing faults.

INVESTIGATION OF THE INFLUENCE OF ADAPTIVE NOISE ON THE EFFICIENCY OF DISCRETE THRESHOLDFUL WAVELET FILTRATION OF MODULATED SIGNALS

To study the filtering of modular signals, a specially generated model signal was used, which consists of attenuating harmonics of a random initial frequency and a phase with half overlap. With the help of the window ganning provides smoothness at the end of the generated signal. Due to the use of Euclidean norms of vectors formed from wavelet coefficients of approximation and detail for all levels of decomposition of the model signal with noise and without obtaining a mathematical model, the change of noise power for filtration efficiency is taken into account. The study of the obtained mathematical model proved the existence of two different levels of decomposition distributed error from the influence of noise for the coefficients of appropriation and detail depending on the number of harmonics in the signal model. For the proposed model signal is indeterminate from the initial frequencies and attenuation coefficients of changes in the error of the error determined through the Euclidean norms of vectors for the coefficients of approximation and detail occurs when the number of harmonics of twenty families is exceeded. An important result of this analysis is that it allows to identify such levels of decomposition at which the relative error of both coefficients of both detail and approximation will be minimal. This confirms the presence of a minimum average dramatic MSE of the threshold filtering efficiency in which remains powerful at the second level of decomposition for the generated model signal when the noise changes. In addition, this error value is further reduced by selecting a discrete wavelet. The non-threshold method of discrete filter wavelet is compared with the methods of general and universal thresholds for the proposed model function in the conditions of noise power change. The unconditional advantage of the method without a threshold is the minimal use of computing resources. The above wavelet filtering technique can be used for other signal functions. The proposed model function is given as a general example.

Application of Wavelet Filtering to Vibrational Signals from the Mining Screen for Spring Condition Monitoring

The main task of mineral processing plants is to further process the raw material extracted in the mining faces into a concentrate with the highest possible concentration of the final product. In practice, it is a complex process in which several stages can be distinguished. After the ore has been transported to the surface by the skip shaft, one of the first steps is sieving the ore, which is typically performed using vibrating mining screens. In a typical Ore Enrichment Plant, the screening process is carried out by several such machines. This is a typical bottleneck in the technological chain. For this reason, the main challenge for users is to achieve the highest reliability and efficiency of these technical facilities. The solution is to focus on predictive maintenance strategies based on the development of monitoring and advanced diagnostic procedures capable of estimating the time of safe operation. This work was developed as part of an advanced diagnostic system ensuring comprehensive technical conditioning and early fault detection of components such as the engine, transmission, bearings, springs, and screen. This article focuses on vibration data. The problem of damage detection in the presence of periodically impulsive components resulting from falling feed material on the screen and its further screening process has been considered. These disturbances are of a non-Gaussian noise nature, the elimination of which is essential to extract the fault-related signal of interest. One solution may be to properly smooth and filter the raw signal. In this article, a wavelet filtering technique is applied. First, the wavelet filtering procedure is described. In the next step, the performance of a wavelet filter is investigated depending on its parameters. Then, the results of wavelet filtering are compared with such methods as low-pass filtering and smoothing using a moving average. Finally, the impact of wavelet filtering on the calculation of screen trajectories is investigated.

A vision-based system for long-distance remote monitoring of dynamic displacement: experimental verification on a supertall structure

Dynamic displacement response of civil structures is an important index for in-construction and in-service structural condition assessment. However, accurately measuring the displacement of large-scale civil structures such as high-rise buildings still remains as a challenging task. In order to cope with this problem, a vision-based system with the use of industrial digital camera and image processing has been developed for long-distance, remote, and real-time monitoring of dynamic displacement of supertall structures. Instead of acquiring image signals, the proposed system traces only the coordinates of the target points, therefore enabling real-time monitoring and display of displacement responses in a relatively high sampling rate. This study addresses the in-situ experimental verification of the developed vision-based system on the Canton Tower of 600 m high. To facilitate the verification, a GPS system is used to calibrate/verify the structural displacement responses measured by the vision-based system. Meanwhile, an accelerometer deployed in the vicinity of the target point also provides frequency-domain information for comparison. Special attention has been given on understanding the influence of the surrounding light on the monitoring results. For this purpose, the experimental tests are conducted in daytime and nighttime through placing the vision-based system outside the tower (in a brilliant environment) and inside the tower (in a dark environment), respectively. The results indicate that the displacement response time histories monitored by the vision-based system not only match well with those acquired by the GPS receiver, but also have higher fidelity and are less noise-corrupted. In addition, the low-order modal frequencies of the building identified with use of the data obtained from the vision-based system are all in good agreement with those obtained from the accelerometer, the GPS receiver and an elaborate finite element model. Especially, the vision-based system placed at the bottom of the enclosed elevator shaft offers better monitoring data compared with the system placed outside the tower. Based on a wavelet filtering technique, the displacement response time histories obtained by the vision-based system are easily decomposed into two parts: a quasi-static ingredient primarily resulting from temperature variation and a dynamic component mainly caused by fluctuating wind load.

The use of MD-CUMSUM and NARX neural network for anticipating the remaining useful life of bearings

The accurate determination of remaining useful life (RUL) of bearings is of immense importance in the condition-based maintenance of any rotating machinery. In this paper, a data driven prognostic approach based on nonlinear autoregressive neural network with eXogenous Inputs (NARX-NN) in combination with wavelet-filter technique is applied to the RUL estimation of bearings. Firstly, the vibration signals generated in an experimental test rig are processed with the proposed wavelet-filter to augment the impulsive characteristics of bearing signals and improve the quality of fault feature extraction. Secondly, a variety of time-domain features are extracted from the processed bearing signals. However, these features exhibit a highly non-monotonic behavior as the bearing condition degrades. To overcome this drawback, a new health indicator (HI) based on Mahalanobis distance (MD) criterion and cumulative sum (CUMSUM) chart is proposed in this paper. Thirdly, the NARX-NN is first designed as a time delay neural network (TDNN). Then, the derived HI and the age of the bearing are used as inputs with life percentage of the bearing as output in order to train the TDNN model, which unlike the usual artificial neural networks (ANNs) performs a one-step ahead prediction of the bearing RUL. The results suggest that the proposed method can effectively predict the RUL of bearings with an acceptable degree of accuracy, and outperforms the use of self-organizing map-based indicator and the traditional FFNNs for RUL inference.

Comment on "Scaling regimes and linear/nonlinear responses of last millennium climate to volcanic and solar forcing" by S. Lovejoy and C. Varotsos (2016)

Abstract. Lovejoy and Varotsos (2016) (L&V) analyse the temperature response to solar, volcanic, and solar plus volcanic forcing in the Zebiak–Cane (ZC) model, and to solar and solar plus volcanic forcing in the Goddard Institute for Space Studies (GISS) E2-R model. By using a simple wavelet filtering technique they conclude that the responses in the ZC model combine subadditively on timescales from 50 to 1000 years. Nonlinear response on shorter timescales is claimed by analysis of intermittencies in the forcing and the temperature signal for both models. The analysis of additivity in the ZC model suffers from a confusing presentation of results based on an invalid approximation, and from ignoring the effect of internal variability. We present tests without this approximation which are not able to detect nonlinearity in the response, even without accounting for internal variability. We also demonstrate that internal variability will appear as subadditivity if it is not accounted for. L&V's analysis of intermittencies is based on a mathematical result stating that the intermittencies of forcing and response are the same if the response is linear. We argue that there are at least three different factors that may invalidate the application of this result for these data. It is valid only for a power-law response function; it assumes power-law scaling of structure functions of forcing as well as temperature signal; and the internal variability, which is strong at least on the short timescales, will exert an influence on temperature intermittence which is independent of the forcing. We demonstrate by a synthetic example that the differences in intermittencies observed by L&V easily can be accounted for by these effects under the assumption of a linear response. Our conclusion is that the analysis performed by L&V does not present valid evidence for a detectable nonlinear response in the global temperature in these climate models.

Wavelet Filter Approach andZ-RRelationship in Meteorological Forecasting

The purpose of this study is to investigate theZ-Rrelationship for computing rainfall using conventional and wavelet filters technique. Wavelet filter technique was applied to data filtration process. The proposed model was applied to determine the rainfall of five rain gauge meteorological stations in Thailand. The three-hourly rainfall and radar reflectivity data were used in this study. The results indicated that the accumulative rainfall of wavelet filters technique was close to the observed rainfall data more than the results of conventional practice for both calibration and validation processes. Consequently, we are confident that a wavelet filters technique is a useful tool for estimating the rainfall.

Multiple dimensioned mining of financial fluctuation through radial basis function networks

Fluctuation mining is one of the greatest challenging tasks in the field of finance market. The main contribution of this research was to propose a multiple dimensioned model for financial market fluctuation mining. In this approach, the original financial time series is broken down into different information by the wavelet filtering technique, and then, all this information is handled through radial basis function networks due to its universal approximation abilities and more robust than the ordinary networks. An experimental analysis is conducted with the proposed model using stock index future time series, revealing consistent performance improvement of this kind of multidimensional approach.

Analysis of Lau interferometric fringe for measurement of focal length by using wavelet filtering techniques

New filtering scheme is investigated and implemented on Lau interferometric fringes to enhance the signal-to-noise ratio (SNR) in the Lau interferograms. Towards this, the applicability of Haar, Daubechies and Symlet wavelet based filtering schemes have been tested. Of these schemes, the Symlet based filtering scheme gives the best results in terms of improved signal to noise ratio and speckle index. This scheme is then used to analyse Lau interferograms recorded in the experimental arrangement for focal length measurement. The processed interferograms yield improved accuracy and precision of measurement.

De-Noising Signal of Electromagnetic Flowmeter Based on Wavelet Transform

A Software filter technique based on Haar wavelet is proposed for effectively suppressing the noise components in the signal of electromagnet flow transducer. It is proved with computer simulation based on MATLAB Wavelet Toolbox that the proposed filter technique is well matched with the feature of the signal of the electromagnetic flow transducer in effectiveness and efficiency. A prototype flowmeter was developed based on the proposed technique, experimental result with the prototype shows that the proposed Haar wavelet filter technique is well suitable for de-noising the signal of electromagnetic flow meter.

Region-Based Fractional Wavelet Transform Using Post Processing Artifact Reduction

Wavelet-based algorithms are increasingly used in the source coding of remote sensing, satellite and other geospatial imagery. At the same time, wavelet-based coding applications are also increased in robust communication and network transmission of images. Although wireless multimedia sensors are widely used to deliver multimedia content due to the availability of inexpensive CMOS cameras, their computational and memory resources are still typically very limited. It is known that allowing a low-cost camera sensor node with limited RAM size to perform a multi-level wavelet transform, will in return limit the size of the acquired image. Recently, fractional wavelet filter technique became an interesting solution to reduce communication energy and wireless bandwidth, for resource-constrained devices (e.g. digital cameras). The reduction in the required memory in these fractional wavelet transforms is achieved at the expense of the image quality. In this paper, an adaptive fractional artifacts reduction approach is proposed for efficient filtering operations according to the desired compromise between the effectiveness of artifact reduction and algorithm simplicity using some local image features to reduce boundaries artifacts caused by fractional wavelet. Applying such technique on different types of images with different sizes using CDF 9/7 wavelet filters results in a good performance.

Robust Prognostics Concept for Gearbox with Artificially Induced Gear Crack Utilizing Acoustic Emission

Prognostic is a rapidly developing field and seeks to build on current diagnostic equipment capabilities for predicting the system state in advance. In machine condition prognostics, the current and past observations are used to predict the upcoming states of the machine. Signal-de-noising and extraction of the weak acoustic signature are crucial to gearbox prognostics since the inherent deficiency of the measuring mechanism often introduces a great amount of noise to the signal. In addition, the signature of a defective gearbox is spread across a wide frequency band and hence can easily become masked by noise and low frequency effects. As a result, robust concepts are needed to provide more evident information for gearbox performance assessment and prognostics. This paper introduces enhanced and robust prognostic concepts for gear tooth based on an optimal wavelet filter method for fault identification and a statistical method for performance degradation assessment. The experimental results demonstrate that the gear tooth defect can be detected and evaluated at an early stage of development when both the optimal wavelet filter and statistical analysis technique are used.

A wavelet filtering based estimation of output gap

Accurate estimation of output gap is an important and challenging problem for any economy. We consider here the problem of estimating the output gap for the Indian economy. The estimation of output gap involves a hypothetical variable, the potential output of the economy. In this paper, we propose a wavelet filtering based technique for estimation of output gap using monthly the Index of Industrial Production (IIP) series. We compare the results obtained using the proposed wavelet based technique with the widely used Hodrick–Prescott filtering and Seasonal ARIMA modeling based techniques. In order to compare the output gap estimated using the proposed technique with the other techniques, we explore the causal relationship between the estimates of output gap and the growth rate of an equity market indicator. It is observed that the wavelet filtering technique gives better results than the more popular econometric and standard time-series modeling techniques.

Low-Memory Wavelet Transforms for Wireless Sensor Networks: A Tutorial

The computational and memory resources of wireless sensor nodes are typically very limited, as the employed low-energy microcontrollers provide only hardware support for 16 bit integer operations and have very limited random access memory (RAM). These limitations prevent the application of modern signal processing techniques to pre-process the collected sensor data for energy and bandwidth efficient transmission over sensor networks. This tutorial introduces communication and networking generalists without a background in wavelet signal processing to low-memory wavelet transform techniques. We first explain the one-dimensional wavelet transform (including the lifting scheme for in-place computation), the two-dimensional wavelet transform, as well as the evaluation of wavelet transforms with fixed-point arithmetic. Then, we explain the fractional wavelet filter technique which computes wavelet transforms with 16 bit integers and requires less than 1.5 kByte of RAM for a 256 × 256 gray scale image. We present case studies illustrating the use of these low-memory wavelet techniques in conjunction with image coding systems to achieve image compression competitive to the JPEG2000 standard on resource-constrained wireless sensor nodes. We make the C-code software for the techniques introduced in this tutorial freely available.

Image Deconvolution Using a General Ridgelet and Curvelet Domain

We carry out deconvolution by transforming the data into a new general discrete Radon domain that can handle any assumed boundary conditions for the associated matrix inversion problem. For each associated component (projection), one can then apply deconvolution routines to smaller (and possibly better) conditioned matrix inversion problems than the matrix inversion problem for the entire image. We demonstrate this new scheme by adaptively deconvolving these components using a combination of regularized inversion and wavelet filtering techniques. This procedure allows us to provide image estimates based on a generalized ridgelet frame. We then devise methods for carrying out this scheme locally to provide estimates based on generalized multiscaled ridgelets which are then filtered and combined to form an estimate from a curvelet-like domain. The techniques presented here suggest a whole new paradigm for developing deconvolution algorithms that incorporate leading deconvolution schemes. Various experimental results show that our methods can perform significantly better than standard deconvolution techniques.

Assessment of glucose metabolism from the projections using the wavelet technique in small animal pet imaging

The dynamic positron emission tomography (PET) images are usually modeled to extract the physiological parameters. However, to avoid reconstruction of the dynamic sequence of images with subjective data filtering, it is advantageous to apply the kinetic modeling in the projection space and to reconstruct single parametric image slices. Using the advantage of the wavelets to compress the data and to filter the noise in the sinogram, we applied the graphical analysis method (Patlak) to generate a single parametric sinogram (WAV-SINO) from PET data acquired in seven normal rats measured with fluorodeoxyglucose (FDG) in the heart. The same data set was analysed with the graphical method in the spatial domain from the sinograms (USUAL-SINO), and also from images reconstructed with non-filtered backprojection (USUAL-nFBP) and filtered backprojection (USUAL-FBP). The myocardial metabolic rates for glucose (MMRG) obtained with USUAL-nFBP, USUAL-FBP, USUAL-SINO and WAV-SINO were found to be, respectively, 7.54, 6.75, 6.52 and 6.98 μmol/100 g/min. While the variance with respect to USUAL-FBP was about 142% for USUAL-nFBP, 99.6% for USUAL-SINO and 101.9% for WAV-SINO, the spatial resolution as assessed from the profiles through the myocardial walls of the reconstructed images was 112% for USUAL-FBP and 105% for WAV-SINO relative to the high resolution USUAL-nFBP. The WAV-SINO parametric images showed slightly better visual quality than those obtained from the spatial domain. Finally, the wavelet filtering technique allowed to reduce the computing time, the storage space and particularly the variance in the MMRG parametric images while preserving the spatial resolution.