Pattern Recognition Algorithms Research Articles

Ensemble machine learning approach for electronic nose signal processing

Electronic nose (e-nose) systems have been reported to be used in many areas as rapid, low-cost, and non-invasive instruments. Especially in meat production and processing, e-nose system is a powerful tool to process volatile compounds as a unique ‘fingerprint’. The ability of the pattern recognition algorithm to analyze e-nose signals is the key to the success of the e-nose system in many applications. On the other hand, ensemble methods have been reported for favorable performances in various data sets. This research proposes an ensemble learning approach for e-nose signal processing, especially in beef quality assessment. Ensemble methods are not only used for learning algorithms but also sensor array optimization. For sensor array optimization, three filter-based feature selection algorithms (FSAs) are used to build ensemble FSA such as reliefF, chi-square, and gini index. Ensemble FSA is developed to deal with different or unstable outputs of a single FSA on homogeneous e-nose data sets in beef quality monitoring. Moreover, ensemble learning algorithms are employed to deal with multi-class classification and regression tasks. Random forest and Adaboost are used that represent bagging and boosting algorithms, respectively. The results are also compared with support vector machine and decision tree as single learners. According to the experimental results, our ensemble approach has good performance and generalization in e-nose signal processing. Optimized sensor combination based on filter-based FSA shows stable results both in classification and regression tasks. Furthermore, Adaboost as a boosting algorithm produces the best prediction even though using a smaller number of sensors.

A training pattern recognition algorithm based on weight clustering for improving cooling load prediction accuracy of HVAC system

The cooling load-based optimal control is an advanced technology for the efficient operation of heating, ventilation, and air conditioning (HVAC). Thus, the prediction reliability of cooling load plays a key role in HVAC's optimal control. Current publications primarily focused on the structure optimization of prediction models, while less on the clustering-based cooling load prediction. However, the data quality determines the upper limit of the model's prediction performance. Thus, a training pattern recognition algorithm based on weight clustering is proposed for improving cooling load prediction accuracy. Compared with the existing clustering-based prediction methods, the main innovations of the proposed method are: (i) considering the input variables' weights on cooling load in the clustering process; and (ii) investigating the matching between the various prediction models and the K-means clustering algorithm. The case studies showed that the proposed method achieves a significant improvement in the prediction performance, such as MAPEs of the MLR, MNR, and ANN decrease by 34.67%, 35.56%, and 14.53% on average, respectively. Compared with the non-weights clustering method, the introduction of the weights can further improve the above models' prediction accuracy, such as their MAPEs decrease by 6.30%, 7.59%, and 3.07% on average, respectively. These results also demonstrated that the clustering-based prediction method is more suitable for the regression models (e.g., MLR and MNR) with low complexity compared to the ANN. When the clustering number is about 4, the models' prediction performances were more robust. Applying the proposed method to the time-series models (i.e., AR, ARX, and ANN) resulted in their MAPEs as low as 1.79%, 1.78%, and 2.06%, respectively. the proposed method can provide a new idea for improving the accuracy of cooling load prediction.

Application of individualized differential expression analysis in human cancer proteome.

Liquid chromatography-mass spectrometry-based quantitative proteomics can measure the expression of thousands of proteins from biological samples and has been increasingly applied in cancer research. Identifying differentially expressed proteins (DEPs) between tumors and normal controls is commonly used to investigate carcinogenesis mechanisms. While differential expression analysis (DEA) at an individual level is desired to identify patient-specific molecular defects for better patient stratification, most statistical DEP analysis methods only identify deregulated proteins at the population level. To date, robust individualized DEA algorithms have been proposed for ribonucleic acid data, but their performance on proteomics data is underexplored. Herein, we performed a systematic evaluation on five individualized DEA algorithms for proteins on cancer proteomic datasets from seven cancer types. Results show that the within-sample relative expression orderings (REOs) of protein pairs in normal tissues were highly stable, providing the basis for individualized DEA for proteins using REOs. Moreover, individualized DEA algorithms achieve higher precision in detecting sample-specific deregulated proteins than population-level methods. To facilitate the utilization of individualized DEA algorithms in proteomics for prognostic biomarker discovery and personalized medicine, we provide Individualized DEP Analysis IDEPAXMBD (XMBD: Xiamen Big Data, a biomedical open software initiative in the National Institute for Data Science in Health and Medicine, Xiamen University, China.) (https://github.com/xmuyulab/IDEPA-XMBD), which is a user-friendly and open-source Python toolkit that integrates individualized DEA algorithms for DEP-associated deregulation pattern recognition.

Application of a Nonparametric Pattern Recognition Algorithm to the Problem of Testing the Hypothesis of the Independence of Variables of Multi-Valued Functions

Application of a Nonparametric Pattern Recognition Algorithm to the Problem of Testing the Hypothesis of the Independence of Variables of Multi-Valued Functions

Real-Time intelligent Elevator Monitoring and Diagnosis: Case Studies and Solutions with applications using Artificial Intelligence

Under "Industry 4.0", the implementation and application of big data and Artificial Intelligence (AI) technology in the elevator industry has become more and more common. With the surge of the elevator operation data and higher requirements for its real-time performance, the traditional elevator fault monitoring is inaccurate, which needs to be solved urgently. In this paper, a fault monitoring and diagnosis method of elevator based on AI and big data is proposed. Firstly, the elevator system and its fault types and causes are analyzed. Then in order to select the best big data processing tools, the performance of Flink and Spark Streaming is compared. The results show that Flink features faster computing speed and is more suitable for handling big data. Thirdly, a pattern recognition algorithm based on finite state machine (FSM) is proposed to monitor the running state for whole elevator control system. Finally, simulation experiment has been made.

Identifying the origins of Tianhuang stones based on Raman spectroscopy and pattern recognition algorithms

In this paper, a non-destructive and efficient method for identifying gemstones of the same species based on Raman spectroscopy and pattern recognition algorithms was developed. Tianhuang stones from different origins (Shoushan (SS), Changhua (CH), and Laos (LA)) with similar composition and structure were investigated and analyzed. Raman spectra were collected by a self-developed portable Raman spectrometer. Combined with pattern recognition methods, the subtle differences between the three types of Tianhuang stones were comprehensively analyzed. Principal component analysis–latent Dirichlet allocation was used to predict the feasibility of distinguishing SS, CH, and LA Tianhuang stones by Raman spectra. Random forest analysis method was applied to establish SS/CH and SS/LA models to distinguish SS from CH and LA. The selection of characteristic variables and determination of the number of growing trees was discussed. Also, the accuracy, sensitivity, and specificity of the two models were calculated. The potential of this method for rapid and non-destructive identification of Tianhuang stones was proved. The feasibility and effectiveness of Raman spectroscopy combined with the pattern recognition method in identifying gemstones of the same species with similar composition were proved.

Gait Analysis for Surveillance

Gait refers to the patterns of movements the limbs make when walking. Human gaits refer to the variety of ways in which an individual can move, either naturally or as a result of training. The gait of each individual is as unique as their voice. On the basis of this knowledge, Machine Learning (ML) algorithms have been developed for gait recognition. Computer Vision (CV) techniques have facilitated the development of a wide range of approaches for identifying people by their movements in videos using both natural biometric characteristics (the human skeleton, silhouette, changes during walking) and abstractions. A gait recognition system identifies the human body based on its shape and the way it moves. A machine-learning system can recognize a person even if their face is hidden, turned away from the camera, or concealed behind a mask. An algorithm analyzes a person's silhouette, height, speed, and walking pattern to identify him or her. Gait recognition technology acquires data from multiple sources, such as video cameras and motion sensors. Data from these sources are then processed by a number of algorithms. Gait is recognized, data is processed, contours and silhouettes are detected, and individual features are segmented, according to the algorithm. After this, the feature extraction algorithm takes effect - this is what differentiates one gait from another. There are many different algorithmic requirements, and these algorithms can vary. Some algorithms, for example, are designed to process video information, while others employ sensor data. Because each gait is distinct, the identification algorithms are always confronted with new data. The system will assess future data better if it detects more gait variants. Assume the program compares two gaits that are highly similar. The algorithms for pattern recognition and silhouette segmentation have been trained to separate the tiny details and enter them into the database. This enables for more accurate gait categorization and improved results in the future.

The finite element method-based pattern recognition approach for the classification of patient-specific gunshot injury.

Violence related injuries and deaths mostly caused by firearms are a major problem throughout the world. Understanding the factors that control the extent of hard-soft tissue wound patterns using computer imaging techniques, numerical methods, and machine learning algorithms may help physicians to diagnose and treat those injuries more properly. Here, we investigate the use of computational results coupled with the pattern recognition algorithms to develop an approach for forensic applications. Initially, computer tomography (CT) images of the patient whose leg was shot by a 9 × 19 parabellum bullet are used to construct the FE models of that patient's femoral bone and the surrounding soft tissues. Then, Hounsfield units-based material properties are assigned to elements of the bone. To simulate the full range of loading conditions encountered in ballistic events, a constitutive model that captures the strain-rate dependent response is implemented. The entrance pathway vector of the bullet is directed in accordance with the patient's wound and the simulations are deployed for the cases having various inlet velocities such as 150, 200, 250, 300, and 350 m/s. Once the FE results for each case are obtained, they are processed with supervised machine learning algorithms to classify the wound and inlet velocity correspondence. The results demonstrate that they can be diagnosed with a percent accuracy of 97.3, 97.5, and 98.3 for the decision tree (DT), k-nearest neighbors (kNN) and support vector machine (SVM) classifier, respectively. This approach may provide a useful framework in classifying the wound type, predicting the bullet impact velocity and its firing distance.

A Novel Similarity Measure of Single-Valued Neutrosophic Sets Based on Modified Manhattan Distance and Its Applications

A single-valued neutrosophic (SVN) set contains three parameters, which can well describe three aspects of an objective thing. However, most previous similarity measures of SVN sets often encounter some counter-intuitive examples. Manhattan distance is a well-known distance, which has been applied in pattern recognition, image analysis, ad-hoc wireless sensor networks, etc. In order to develop suitable distance measures, a new distance measure of SVN sets based on modified Manhattan distance is constructed, and a new distance-based similarity measure also is put forward. Then some applications of the proposed similarity measure are introduced. First, we introduce a pattern recognition algorithm. Then a multi-attribute decision-making method is proposed, in which a weighting method is developed by building an optimal model based on the proposed similarity measure. Furthermore, a clustering algorithm is also put forward. Some examples are also used to illustrate these methods.

Naive Bayesian Prediction of Japanese Annotated Corpus for Textual Semantic Word Formation Classification

With the rapid development of Japanese information processing technology, problems such as polysemy and ambiguity at the text and dialogue level, as well as unregistered words, have become increasingly prominent because computers cannot fully “understand” the semantics of words. How to make the computer “understand” the semantics of words accurately requires the computer to “understand” the rules of converting and integrating words into words from the perspective of semantics. Traditional Japanese text classification mostly adopts the text representation method of vector space model, which has the problem of confusing classification effect. Therefore, this paper proposes the topic of constructing a semantic word formation pattern prediction model based on a large-scale annotated corpus. This paper proposes a solution that combines Japanese semantic word formation rules with pattern recognition algorithms. Aiming at this scheme, a variety of pattern recognition algorithms were compared and analyzed, and the naive Bayesian model was decided to predict semantic word formation patterns. This paper further improves the accuracy of computer prediction of Japanese semantic word formation patterns by adding part of speech. Before modeling, the parts of speech of words are automatically tagged and manually checked based on the original annotated corpus. In the research on predicting Japanese semantic word formation patterns, this paper builds a semantic word formation pattern prediction model based on Naive Bayes and conducts simulation experiments. We divide the eight types of semantic word formation patterns in the annotated corpus into two groups, and divide the obtained sample sets into training sets and test sets, so that the Naive Bayes model first learns semantic word formation rules based on the training sets of each group. Semantic word formation patterns are predicted on the test set for each group. The simulation results show that the prediction model of semantic word formation mode has a generally high degree of fit and prediction accuracy. The prediction model of semantic word formation pattern based on this theory can ensure that the computer can judge the semantic word formation pattern more accurately.

A new distance measure for pythagorean fuzzy sets based on earth mover’s distance and its applications

In uncertain information processing, new knowledge can be discovered by measuring the proximity between discovered and undiscovered knowledge. Pythagorean Fuzzy Sets (PFSs) is one of the important tools to describe the natural attributes of uncertain information. Therefore, how to appropriately measure the distance between PFSs is an important topic. The earth mover’s distance (EMD) is a real distance metric that can be used to describe the difference between two distribution laws. In this paper, a new distance measure for PFSs based on EMD is proposed. It is a new perspective to measure the distance between PFSs from the perspective of distribution law. First, a new distance measure namely DEMD is presented and proven to satisfy the distance measurement axiom. Second, an example is given to illustrate the advantages of DEMD compared with other distance measures. Third, the problem statements and solving algorithms of pattern recognition, medical diagnosis and multi-criteria decision making (MCDM) problems are given. Finally, by comparing the application of different methods in pattern recognition, medical diagnosis and MCDM, the effectiveness and practicability of DEMD and algorithms presented in this paper are demonstrated.

Complex belief interval‐based distance measure with its application in pattern recognition

The complex evidence theory is an effective methodology for multiattribute decision-making. Since difference measure between multiattribute plays an important role for conflict management in the process of multiattribute decision-making, how to measure discrepancy between complex basic belief assignments (CBBAs) in complex evidence theory is still an open issue. In this context, a new distance measurement (complex belief distance—CBD) is proposed in this paper by taking advantages of complex belief function and complex plausibility function, called complex belief interval-based distance. In addition, we compare the proposed CBD with the related work to illustrate its superiority. Next, based on CBD, we devise a novel multiattribute decision-making algorithm for pattern recognition. Finally, we apply the method to problems of medical diagnosis to verify the effectiveness of the proposed method.

Information system of remote sensing data analysis

The information system is designed to solve the problems of analyzing remote sensing data and making decisions using non-parametric pattern recognition algorithms, the synthesis of which is based on kernel estimates of probability density. Initial information is compiled by training samples on the states of the investigated natural objects and their corresponding spectral data, which are presented in Microsoft Excel files. The application of the software provides the following functions: primary processing of remote sensing data to evaluate the main quantitative characteristics of the laws of distribution of random variables; testing of the hypothesis on distributions of multivariate random variables based on non-parametric pattern recognition algorithms; estimating the state of natural objects by spectral data; mapping of territories by state of natural objects based on spectral data of earth surface elements and their corresponding spatial coordinates. Results of the presented functional tasks are saved in the form of reports in graphic files and spreadsheets of Microsoft Excel. Software tools can be used in the study of various nature objects from remote sensing data.

Hough Transform Proposal and Simulations for Particle Track Recognition for LHC Phase-II Upgrade

In the near future, LHC experiments will continue future upgrades by overcoming the technological obsolescence of the detectors and the readout capabilities. Therefore, after the conclusion of a data collection period, CERN will have to face a long shutdown to improve overall performance, by updating the experiments, and implementing more advanced technologies and infrastructures. In particular, the largest LHC experiment, i.e., ATLAS, will upgrade parts of the detector, the trigger, and the data acquisition system. In addition, the ATLAS experiment will complete the implementation of new strategies, algorithms for data handling, and transmission to the final storage apparatus. This paper presents an overview of an upgrade planned for the second half of this decade for the ATLAS experiment. In particular, we show a study of a novel pattern recognition algorithm used in the trigger system, which is a device designed to provide the information needed to select physical events from unnecessary background data. The idea is to use a well known mathematical transform, the Hough transform, as the algorithm for the detection of particle trajectories. The effectiveness of the algorithm has already been validated in the past, regardless of particle physics applications, to recognize generic shapes within images. On the contrary, here, we first propose a software emulation tool, and a subsequent hardware implementation of the Hough transform, for particle physics applications. Until now, the Hough transform has never been implemented on electronics in particle physics experiments, and since a hardware implementation would provide benefits in terms of overall Latency, we complete the studies by comparing the simulated data with a physical system implemented on a Xilinx hardware accelerator (FELIX-II card). In more detail, we have implemented a low-abstraction RTL design of the Hough transform on Xilinx UltraScale+ FPGAs as target devices for filtering applications.

Prediction of Chinese Semantic Word-Building Patterns Based on Complex Network Features

The research on prediction of Chinese semantic word-formation patterns based on complex network features has certain practical and theoretical significance in the field of natural language understanding. In this paper, complex networks are introduced into the prediction of Chinese semantic word-formation patterns, and a new prediction method of Chinese semantic word-formation patterns based on complex networks is proposed. And a solution that combines the semantic word-building rules of Chinese language with pattern recognition algorithm is put forward. Aiming at this scheme, a variety of pattern recognition algorithms are compared and analyzed, and the most suitable binary logistic regression model and naive Bayes model are found to predict Chinese semantic word-building patterns. The semantic loss is reduced, and the text classification model and corresponding classification algorithm are constructed, by introducing the maximum common subgraph theory to calculate text similarity under the complex network representation. The results of the experiments show that using complex networks to predict Chinese semantic word-formation patterns is both effective and feasible. The computer can judge the semantic word-formation pattern more accurately using the semantic word-formation pattern prediction model based on this theory.

Intelligent fault diagnosis of train axle box bearing based on parameter optimization VMD and improved DBN

The vibration signal of the axle box bearing of the train is affected by the track excitation and the random noise of the environment. The vibration signal is nonlinear and non-stationary, and the signal characteristics of the early fault are weak and easy to be submerged, which leads to the low accuracy of the weak fault diagnosis of the bearing. To solve this problem, a weak fault diagnosis method for train axle box bearing based on parameter optimization Variational Mode Decomposition (VMD) and improved Deep Belief Network (DBN) is proposed. Firstly, the nonlinear convergence factor, Levy flight theory and greedy algorithm optimization theory are introduced into the Grey Wolf optimization algorithm (GWO), and an improved GWO algorithm based on hybrid strategy is proposed to improve the performance of the algorithm and solve the local optimal problem of the algorithm. Secondly, the improved GWO is applied to optimize the VMD parameters, which is used for signal decomposition. And the fault feature information of modal components with maximum correlation coefficient is extracted by multi-scale scatter entropy. Finally, the improved GWO algorithm is applied to optimize the parameters of the DBN to solve the parameter setting problem, and the optimized DBN is used as a pattern recognition algorithm for weak fault diagnosis of bearings. Through experimental comparison and analysis, the proposed method can effectively solve the problem of weak fault diagnosis of axle box bearings, and has high diagnostic accuracy.

Dissimilarity-vector spaces based on Dynamic Time Warpings of spectral/time-frequency information for structural health monitoring

This paper presents a powerful data representation, so far unexplored for data-based Structural Health Monitoring, relying on the dissimilarity pattern recognition paradigm and the proximity-learning, providing highly discriminant dissimilarity-based vector spaces, —also called generalized dissimilarity kernels—, where any classifier can be trained for damage classification issues. Conventionally, these damage detection tasks involve a domain-dependent and, quite often, non-trivial preprocessing step by which is computed a feature set from each observation. A crucial consequence is that valuable structural information could be lost in this feature extraction step, leading to models with poor performance. In particular, in this paper we introduce a novel type of dissimilarity-based vector spaces for structural health diagnosis, building up them on a direct pairwise comparison between spectral/time–frequency structural information via the Dynamic Time Warping distance, without a previous feature extraction step, for learning one-class classifiers and using only undamaged data during training. The very sound results, using two data sets widely referenced in the scientific literature, clearly show its potential to complement the state-of-art of pattern recognition algorithms that are used on data-based Structural Health Monitoring.

Digital PCR system development accelerator—A methodology to emulate dPCR results

The development of a digital polymerase chain reaction (dPCR) system typically begins with an idea for the system configuration and chip layout, followed by design, fabrication, and hardware testing. The image processing software can be developed and verified based on the test results. In this paper, we proposed a dPCR emulation methodology to train the developed image processing software before building the dPCR system hardware. We developed a script in a MATLAB environment to generate artificial dPCR images and emulate the dPCR results. First, we defined the number of parameters corresponding to the emulated results, such as the number of partitions with targets, background fluorescence distribution and intensity, image defects, image rotation angle, shift, non-uniform light distribution, and temperature sensitivity. We then implemented the defined parameters and generated an artificial dPCR chip image based on layout design or pattern recognition algorithm. Finally, we obtained a dataset from the artificial image for subsequent result analysis. The generated images could then be used to train the image processing algorithms based on the requirements. We verified the proposed method using various designs of dPCR chips from recently published papers, demonstrating the method's versatility. The proposed method also demonstrated the capability for separating the software and hardware development. Thus, our method allowed the image processing and hardware to be concurrently designed and tested simplifying and speeding up the dPCR system development.

Identification of Chemical Vapor Mixture Assisted by Artificially Extended Database for Environmental Monitoring.

A fully integrated sensor array assisted by pattern recognition algorithm has been a primary candidate for the assessment of complex vapor mixtures based on their chemical fingerprints. Diverse prototypes of electronic nose systems consisting of a multisensory device and a post processing engine have been developed. However, their precision and validity in recognizing chemical vapors are often limited by the collected database and applied classifiers. Here, we present a novel way of preparing the database and distinguishing chemical vapor mixtures with small data acquisition for chemical vapors and their mixtures of interest. The database for individual vapor analytes is expanded and the one for their mixtures is prepared in the first-order approximation. Recognition of individual target vapors of NO2, HCHO, and NH3 and their mixtures was evaluated by applying the support vector machine (SVM) classifier in different conditions of temperature and humidity. The suggested method demonstrated the recognition accuracy of 95.24%. The suggested method can pave a way to analyze gas mixtures in a variety of industrial and safety applications.

Symmetrically Stacked Long Short-Term Memory Networks for Fall Event Recognition Using Compact Convolutional Neural Networks-Based Tracker

In recent years, the advancement of pattern recognition algorithms, specifically the deep learning-related techniques, have propelled a tremendous amount of researches in fall event recognition systems. It is important to detect a fall incident as early as possible, whereby a slight delay in providing immediate assistance can cause severe unrecoverable injuries. One of the main challenges in fall event recognition is the imbalanced training data between fall and no-fall events, where a real-life fall incident is a sporadic event that occurs infrequently. Most of the recent techniques produce a lot of false alarms, as it is hard to train them to cover a wide range of fall situations. Hence, this paper aims to detect the exact fall frame in a video sequence, as such it will not be dependent on the whole clip of the video sequence. Our proposed approach consists of a two-stage module where the first stage employs a compact convolutional neural network tracker to generate the object trajectory information. Features of interest will be sampled from the generated trajectory paths, which will be fed as the input to the second stage. The next stage network then models the temporal dependencies of the trajectory information using symmetrical Long Short-Term Memory (LSTM) architecture. This two-stage module is a novel approach as most of the techniques rely on the detection module rather than the tracking module. The simulation experiments were tested using Fall Detection Dataset (FDD). The proposed approach obtains an expected average overlap of 0.167, which is the best performance compared to Multi-Domain Network (MDNET) and Tree-structured Convolutional Neural Network (TCNN) trackers. Furthermore, the proposed 3-layers of stacked LSTM architecture also performs the best compared to the vanilla recurrent neural network and single-layer LSTM. This approach can be further improved if the tracker model is firstly pre-tuned in offline mode with respect to a specific type of object of interest, rather than a general object.