Pattern Recognition Scheme Research Articles

Modern Computer Vision Techniques for X-Ray Testing in Baggage Inspection

X-ray screening systems have been used to safeguard environments in which access control is of paramount importance. Security checkpoints have been placed at the entrances to many public places to detect prohibited items, such as handguns and explosives. Generally, human operators are in charge of these tasks as automated recognition in baggage inspection is still far from perfect. Research and development on X-ray testing is, however, exploring new approaches based on computer vision that can be used to aid human operators. This paper attempts to make a contribution to the field of object recognition in X-ray testing by evaluating different computer vision strategies that have been proposed in the last years. We tested ten approaches. They are based on bag of words, sparse representations, deep learning, and classic pattern recognition schemes among others. For each method, we: 1) present a brief explanation; 2) show experimental results on the same database; and 3) provide concluding remarks discussing pros and cons of each method. In order to make fair comparisons, we define a common experimental protocol based on training, validation, and testing data (selected from the public ${\mathbb {GDX}}$ ray database). The effectiveness of each method was tested in the recognition of three different threat objects: 1) handguns; 2) shuriken (ninja stars); and 3) razor blades. In our experiments, the highest recognition rate was achieved by methods based on visual vocabularies and deep features with more than 95% of accuracy. We strongly believe that it is possible to design an automated aid for the human inspection task using these computer vision algorithms.

Double-Barrier Memristive Devices for Unsupervised Learning and Pattern Recognition.

The use of interface-based resistive switching devices for neuromorphic computing is investigated. In a combined experimental and numerical study, the important device parameters and their impact on a neuromorphic pattern recognition system are studied. The memristive cells consist of a layer sequence Al/Al2O3/NbxOy/Au and are fabricated on a 4-inch wafer. The key functional ingredients of the devices are a 1.3 nm thick Al2O3 tunnel barrier and a 2.5 mm thick NbxOy memristive layer. Voltage pulse measurements are used to study the electrical conditions for the emulation of synaptic functionality of single cells for later use in a recognition system. The results are evaluated and modeled in the framework of the plasticity model of Ziegler et al. Based on this model, which is matched to experimental data from 84 individual devices, the network performance with regard to yield, reliability, and variability is investigated numerically. As the network model, a computing scheme for pattern recognition and unsupervised learning based on the work of Querlioz et al. (2011), Sheridan et al. (2014), Zahari et al. (2015) is employed. This is a two-layer feedforward network with a crossbar array of memristive devices, leaky integrate-and-fire output neurons including a winner-takes-all strategy, and a stochastic coding scheme for the input pattern. As input pattern, the full data set of digits from the MNIST database is used. The numerical investigation indicates that the experimentally obtained yield, reliability, and variability of the memristive cells are suitable for such a network. Furthermore, evidence is presented that their strong I–V non-linearity might avoid the need for selector devices in crossbar array structures.

A Novel Fiber Optic Based Surveillance System for Prevention of Pipeline Integrity Threats.

This paper presents a novel surveillance system aimed at the detection and classification of threats in the vicinity of a long gas pipeline. The sensing system is based on phase-sensitive optical time domain reflectometry (ϕ-OTDR) technology for signal acquisition and pattern recognition strategies for threat identification. The proposal incorporates contextual information at the feature level and applies a system combination strategy for pattern classification. The contextual information at the feature level is based on the tandem approach (using feature representations produced by discriminatively-trained multi-layer perceptrons) by employing feature vectors that spread different temporal contexts. The system combination strategy is based on a posterior combination of likelihoods computed from different pattern classification processes. The system operates in two different modes: (1) machine + activity identification, which recognizes the activity being carried out by a certain machine, and (2) threat detection, aimed at detecting threats no matter what the real activity being conducted is. In comparison with a previous system based on the same rigorous experimental setup, the results show that the system combination from the contextual feature information improves the results for each individual class in both operational modes, as well as the overall classification accuracy, with statistically-significant improvements.

Complex adaptive systems and their relevance for nursing: An evolutionary concept analysis.

This study aimed to analyse the concept of "complex adaptive systems." The construct is still nebulous in the literature, and a further explanation of the idea is needed to have a shared knowledge of it. A concept analysis was conducted utilizing Rodgers evolutionary method. The inclusive years of bibliographic search started from 2005 to 2015. The search was conducted at PubMed©, CINAHL© (EBSCO host©), Scopus©, Web of Science©, and Academic Search Premier©. Retrieved papers were critically analysed to explore the attributes, antecedents, and consequences of the concept. Moreover, surrogates, related terms, and a pattern recognition scheme were identified. The concept analysis showed that complex systems are adaptive and have the ability to process information. They can adapt to the environment and consequently evolve. Nursing is a complex adaptive system, and the nursing profession in practice exhibits complex adaptive system characteristics. Complexity science through complex adaptive systems provides new ways of seeing and understanding the mechanisms that underpin the nursing profession.

Cosmic-Ray Extremely Distributed Observatory: a global cosmic ray detection framework

The main objective of the Cosmic-Ray Extremely Distributed Observatory (CREDO) is the detection and analysis of extended cosmic ray phenomena, so-called super-preshowers (SPS), using existing as well as new infrastructure (cosmic-ray observatories, educational detectors, single detectors etc.). The search for ensembles of cosmic ray events initiated by SPS is yet an untouched ground, in contrast to the current state-of-the-art analysis, which is focused on the detection of single cosmic ray events. Theoretical explanation of SPS could be given either within classical (e.g., photon-photon interaction) or exotic (e.g., Super Heavy Dark Matter decay or annihilation) scenarios, thus detection of SPS would provide a better understanding of particle physics, high energy astrophysics and cosmology. The ensembles of cosmic rays can be classified based on the spatial and temporal extent of particles constituting the ensemble. Some classes of SPS are predicted to have huge spatial distribution, a unique signature detectable only with a facility of the global size. Since development and commissioning of a completely new facility with such requirements is economically unwarranted and time-consuming, the global analysis goals are achievable when all types of existing detectors are merged into a worldwide network. The idea to use the instruments in operation is based on a novel trigger algorithm: in parallel to looking for neighbour surface detectors receiving the signal simultaneously, one should also look for spatially isolated stations clustered in a small time window. On the other hand, CREDO strategy is also aimed at an active engagement of a large number of participants, who will contribute to the project by using common electronic devices (e.g., smartphones), capable of detecting cosmic rays. It will help not only in expanding the geographical spread of CREDO, but also in managing a large manpower necessary for a more efficient crowd-sourced pattern recognition scheme to identify and classify SPS. A worldwide network of cosmic-ray detectors could not only become a unique tool to study fundamental physics, it will also provide a number of other opportunities, including space-weather or geophysics studies. Among the latter one has to list the potential to predict earthquakes by monitoring the rate of low energy cosmic-ray events. The diversity of goals motivates us to advertise this concept across the astroparticle physics community.

Quality control in hard disc drive manufacturing using pattern recognition technique

Computerized monitoring-diagnosis is an efficient technique to identify the source of unnatural variation (UV) in manufacturing process. In this study, a pattern recognition scheme (PRS) for monitoring-diagnosis the UVs was developed based on control chart pattern recognition technique. This PRS integrates the multivariate exponentially weighted moving average (MEWMA) control chart and artificial neural network (ANN) recognizer to perform two-stage monitoring-diagnosis. The first stage monitoring was performed using the MEWMA statistics, whereas the second stage monitoring-diagnosis was performed using an ANN. The PRS was designed based on bivariate process mean shifts between 0.75σ and 3.00σ, with cross correlation between ρ=0.1 and 0.9. The performance of the proposed PRS has been validated in quality control of hard disk drive component manufacturing. The validation proved that it is efficient in rapidly detecting UV and accurately classify the source of UV patterns. In a nutshell, the PRS will aid in realizing automated decision making system in manufacturing industry.

Self-learning kinetic Monte Carlo simulations of Al diffusion in Mg

Vacancy-mediated diffusion of an Al atom in the pure Mg matrix is studied using the atomistic, on-lattice self-learning kinetic Monte Carlo (SLKMC) method. Activation barriers for vacancy-Mg and vacancy-Al atom exchange processes are calculated on the fly using the climbing image nudged-elastic-band method and binary Mg–Al modified embedded-atom method interatomic potential. Diffusivities of an Al atom obtained from SLKMC simulations show the same behavior as observed in experimental and theoretical studies available in the literature; that is, an Al atom diffuses faster within the basal plane than along the c-axis. Although the effective activation barriers for an Al atom diffusion from SLKMC simulations are close to experimental and theoretical values, the effective prefactors are lower than those obtained from experiments. We present all the possible vacancy-Mg and vacancy-Al atom exchange processes and their activation barriers identified in SLKMC simulations. A simple mapping scheme to map an HCP lattice onto a simple cubic lattice is described, which enables simulation of the HCP lattice using the on-lattice framework. We also present the pattern recognition scheme which is used in SLKMC simulations to identify the local Al atom configuration around a vacancy.

Toward Prevention of Pipeline Integrity Threats Using a Smart Fiber-Optic Surveillance System

This paper presents the first available report in the literature of a system aimed at the detection and classification of threats in the vicinity of a long gas pipeline. The system is based on phase-sensitive optical time-domain reflectometry technology for signal acquisition and pattern recognition strategies for threat identification. The system operates in two different modes: 1) machine+activity identification, which outputs the activity being carried out by a certain machine; and 2) threat detection, aimed at detecting threats no matter what the real activity being conducted is. Different strategies dealing with position selection and normalization methods are presented and evaluated using a rigorous experimental procedure on realistic field data. Experiments are conducted with eight machine+activity pairs, which are further labeled as threat or nonthreat for the second mode of the system. The results obtained are promising given the complexity of the task and open the path to future improvements toward fully functional pipeline threat detection systems operating in real conditions.

Self Absorption: Where will technology lead us?

Looking back on my early experiences as a young engineer, I am reminded how little my colleagues and I appreciated that what we did would change the world, for better and for worse. I am also reminded how Marcel Golay, one of my early mentors, understood the duality of technology and how it plays a large part in its application for the right purpose. Born in Switzerland in 1902, Golay received his engineering degree from the Swiss Federal Institute of Technology in Zurich, the same college from which Einstein graduated. He is best known for the invention of the Golay cell used in gas chromatography and optical spectroscopy, the development of the Savitzky?Golay smoothing filter, the discovery of Golay codes, the generalization of the perfect binary Hamming codes to nonbinary codes, and his work on the complementary sequence autocorrelation functions used in Wi-Fi and 3G standards and on the Golay hexagonal nearest-neighbor logic used in pattern recognition schemes, of which I had the good fortune to be a part.

Oblique rotation of factors: a novel pattern recognition strategy to classify fluorescence excitation-emission matrices of human blood plasma for early diagnosis of colorectal cancer.

Colorectal cancer (CRC) ranks high in both men and women, accounting for about 13% of all cancers. In this study, a novel pattern recognition strategy is proposed to improve early diagnosis of CRC through visualizing the relationship between different spectral patterns in a case-control research. Partial least squares-discriminant analysis (PLS-DA) and supervised Kohonen network (SKN) were used to classify the fluorescence excitation-emission matrices (EEMs) from 289 human blood plasma samples containing CRC patients, adenomas tumor, other non-malignant findings and healthy individuals. To obtain optimal factors, oblique rotation (OR) and genetic algorithm (GA) were used to rotate the factors by optimizing transformation matrix elements. Transformed factors were introduced to SKN to build a classification model and the model performance was examined via comparison with a common classifier; PLS-DA. Classification models were built for CRC-healthy and adenomas-healthy samples and the best results were obtained through applying GA-OR on PLS factors and introducing them to the classifiers. Non-error rates for SKN and PLS-DA models assisted with GA (for selecting more informative PLS factors) and OR were equal to 0.97 and 0.95 in cross validation and 0.93 and 0.90 for prediction of the external test set, respectively. Moreover, according to the acceptable results for adenomas-healthy cases using optimal factors, CRC can be diagnosed in early stages. Combining classifiers and optimal factors proved to be efficient for distinguishing healthy and malignant samples, and OR can significantly improve performance of the classification model.

Fast and automatic microseismic phase-arrival detection and denoising by pattern recognition and reduced-rank filtering

We have developed a fast method that allowed us to automatically detect and denoise microseismic phase arrivals from 3C multichannel data. The method is a two-step process. First, the detection is carried out by means of a pattern recognition strategy that seeks plausible hyperbolic phase arrivals immersed in noisy 3C multichannel data. Then, the microseismic phase arrivals are denoised and reconstructed using a reduced-rank approximation of the singular value decomposition of the data along the detected phase arrivals in the context of a deflation procedure that took into account multiple arrivals and/or phases. For the detection, we have defined an objective function that measured the energy and coherence of a potential microseismic phase arrival along an apex-shifted hyperbolic search window. The objective function, which was maximized using very fast simulated annealing, was based on the energy of the average signal and depended on the source position, receivers geometry, and velocity. In practice, the detection process did not require any a priori velocity model, leading to a fast algorithm that can be used in real time, even when the underlying velocity model was not constant. The reduced-rank filtering coupled with a crosscorrelation-based synchronization strategy allowed us to extract the most representative waveform for all the individual traces. Tests using synthetic and field data have determined the reliability and effectiveness of the proposed method for the accurate detection and denoising of 3C multichannel microseismic events under noisy conditions. Two confidence indicators to assess the presence of an actual phase arrival and the reliability of the denoised individual wave arrivals were also developed.

Detection of Alzheimer's disease signature in MR images seven years before conversion to dementia: Toward an early individual prognosis.

Finding very early biomarkers of Alzheimer's Disease (AD) to aid in individual prognosis is of major interest to accelerate the development of new therapies. Among the potential biomarkers, neurodegeneration measurements from MRI are considered as good candidates but have so far not been effective at the early stages of the pathology. Our objective is to investigate the efficiency of a new MR-based hippocampal grading score to detect incident dementia in cognitively intact patients. This new score is based on a pattern recognition strategy, providing a grading measure that reflects the similarity of the anatomical patterns of the subject under study with dataset composed of healthy subjects and patients with AD. Hippocampal grading was evaluated on subjects from the Three-City cohort, with a followup period of 12 years. Experiments demonstrate that hippocampal grading yields prediction accuracy up to 72.5% (P < 0.0001) 7 years before conversion to AD, better than both hippocampal volume (58.1%, P = 0.04) and MMSE score (56.9%, P = 0.08). The area under the ROC curve (AUC) supports the efficiency of imaging biomarkers with a gain of 8.4 percentage points for hippocampal grade (73.0%) over hippocampal volume (64.6%). Adaptation of the proposed framework to clinical score estimation is also presented. Compared with previous studies investigating new biomarkers for AD prediction over much shorter periods, the very long followup of the Three-City cohort demonstrates the important clinical potential of the proposed imaging biomarker. The high accuracy obtained with this new imaging biomarker paves the way for computer-based prognostic aides to help the clinician identify cognitively intact subjects that are at high risk to develop AD.

Validation of a Novel Digital Tool in Automatic Scoring of an Online ECG Examination at an International Cardiology Meeting.

We have previously developed a novel digital tool capable of automatically recognizing correct electrocardiography (ECG) diagnoses in an online exam and demonstrated a significant improvement in diagnostic accuracy when utilizing an inductive-deductive reasoning strategy over a pattern recognition strategy. In this study, we sought to validate these findings from participants at the International Winter Arrhythmia School meeting, one of the foremost electrophysiology events in Canada. Preregistration to the event was sent by e-mail. The exam was administered on day 1 of the conference. Results and analysis were presented the following morning to participants. Twenty-five attendees completed the exam, providing a total of 500 responses to be marked. The online tool accurately identified 195 of a total of 395 (49%) correct responses (49%). In total, 305 responses required secondary manual review, of which 200 were added to the correct responses pool. The overall accuracy of correct ECG diagnosis for all participants was 69% and 84% when using pattern recognition or inductive-deductive strategies, respectively. Utilization of a novel digital tool to evaluate ECG competency can be set up as a workshop at international meetings or educational events. Results can be presented during the sessions to ensure immediate feedback.

Blended Power Management Strategy Using Pattern Recognition for a Plug-in Hybrid Electric Vehicle

The dual power source of a plug-in hybrid electric vehicle (PHEV) requires a high level control strategy in order to establish a power split decision that will minimize fuel consumption while taking full advantage of the embedded source of electrical energy. Literature shows that the optimal control of the power split is greatly influenced by the future trip to be made and that blended strategies are more appropriate regarding battery usage throughout a trip. This paper proposes a blended strategy for a PHEV which uses a driving pattern recognition scheme that allows control adaptation in real-time regarding current driving conditions.

From micropterism to hyperpterism: recognition strategy and standardized homology-driven terminology of the forewing venation patterns in planthoppers (Hemiptera: Fulgoromorpha).

Following recent advances in the morphological interpretations of the tegmen basal cell margins in the Paraneoptera, a standardized and homology-driven groundplan terminology for tegmina types, structures and vein patterns in Hemiptera Fulgoromorpha, including fossils, is proposed. Each term is listed with a morphological definition, compared and linked to the main systems of planthopper forewing description that have been reviewed. The importance of a standardized and homology-driven terminology is stressed to enhance the quality of data in taxonomic descriptions and to strengthen phylogenetic morphological analysis results. When the interpretation of the origin of vein branches is render difficult, a three-step strategy for pattern recognition of the vein is proposed based on two principles: (1) vein forks are more informative than topology of the vein branches: a search for homologous areas, the nodal cells in particular, must first guide the recognition rather the number of branches of a vein, and (2) minimum of ad hoc evolutionary events should be invoked in the understanding of a modified vein pattern. Examples of some conflicting interpretations of venation patterns in planthoppers are discussed within different families for both extant and extinct taxa. For the first time, the concept of brachypterism is defined in a non-relative way independently from other structures, and the new one of hyperpterism is proposed; a reporting system is proposed for each of them.

Pattern recognition based on the correlated intensity fluctuations of thermal light.

Here we present a pattern recognition scheme based on the intensity correlation of thermal light. We prove theoretically that under spatially incoherent illumination the matched filtering technique can be realized in the ghost imaging field. Using the matched filtering technique, it is possible to distinguish an object from a preestablished set of objects through their ghost images, which are extracted by means of intensity correlation measurement. According to the pattern recognition scheme, we present a numerical simulation in which we can easily identify the character inserted into the object arm from a set of two characters through the position of the autocorrelation peak. This pattern recognition scheme opens up the possibility of performing coherent optical processing under spatially incoherent illumination.

Scattering features for lung cancer detection in fibered confocal fluorescence microscopy images.

To assess the feasibility of lung cancer diagnosis using fibered confocal fluorescence microscopy (FCFM) imaging technique and scattering features for pattern recognition. FCFM imaging technique is a new medical imaging technique for which interest has yet to be established for diagnosis. This paper addresses the problem of lung cancer detection using FCFM images and, as a first contribution, assesses the feasibility of computer-aided diagnosis through these images. Towards this aim, we have built a pattern recognition scheme which involves a feature extraction stage and a classification stage. The second contribution relies on the features used for discrimination. Indeed, we have employed the so-called scattering transform for extracting discriminative features, which are robust to small deformations in the images. We have also compared and combined these features with classical yet powerful features like local binary patterns (LBP) and their variants denoted as local quinary patterns (LQP). We show that scattering features yielded to better recognition performances than classical features like LBP and their LQP variants for the FCFM image classification problems. Another finding is that LBP-based and scattering-based features provide complementary discriminative information and, in some situations, we empirically establish that performance can be improved when jointly using LBP, LQP and scattering features. In this work we analyze the joint capability of FCFM images and scattering features for lung cancer diagnosis. The proposed method achieves a good recognition rate for such a diagnosis problem. It also performs well when used in conjunction with other features for other classical medical imaging classification problems.

A multi-stage control chart pattern recognition scheme based on independent component analysis and support vector machine

Recognition of unnatural control chart patterns (CCPs) is an important issue because the unnatural CCPs can be associated with specific assignable causes negatively affecting the manufacturing process. By assuming that an unnatural CCP is a combination of normal pattern and process disturbance, a multi-stage control chart pattern recognition scheme which integrates independent component analysis (ICA) and support vector machine (SVM) is proposed in this study. The proposed multi-stage ICA-SVM scheme first uses ICA to extract independent components (ICs) from the monitoring process data containing CCPs. The normal pattern and process disturbance hidden in the process data can be discovered in the ICs. Then, the IC representing the process disturbance can be identified. Finally, the identified IC and the data of the monitoring process are used as input variables to develop three different SVM models for CCP recognition. The simulation results show that the proposed multi-stage ICA-SVM scheme not only produces accurate and stable recognition results but also has better classification accuracy than four competing models.

Identification of Natural Metabolites in Mixture: A Pattern Recognition Strategy Based on 13C NMR

Because of their highly complex metabolite profile, the chemical characterization of bioactive natural extracts usually requires time-consuming multistep purification procedures to achieve the structural elucidation of pure individual metabolites. The aim of the present work was to develop a dereplication strategy for the identification of natural metabolites directly within mixtures. Exploiting the polarity range of metabolites, the principle was to rapidly fractionate a multigram quantity of a crude extract by centrifugal partition extraction (CPE). The obtained fractions of simplified chemical composition were subsequently analyzed by (13)C NMR. After automatic collection and alignment of (13)C signals across spectra, hierarchical clustering analysis (HCA) was performed for pattern recognition. As a result, strong correlations between (13)C signals of a single structure within the mixtures of the fraction series were visualized as chemical shift clusters. Each cluster was finally assigned to a molecular structure with the help of a locally built (13)C NMR chemical shift database. The proof of principle of this strategy was achieved on a simple model mixture of commercially available plant secondary metabolites and then applied to a bark extract of the African tree Anogeissus leiocarpus Guill. & Perr. (Combretaceae). Starting from 5 g of this genuine extract, the fraction series was generated by CPE in only 95 min. (13)C NMR analyses of all fractions followed by pattern recognition of (13)C chemical shifts resulted in the unambiguous identification of seven major compounds, namely, sericoside, trachelosperogenin E, ellagic acid, an epimer mixture of (+)-gallocatechin and (-)-epigallocatechin, 3,3'-di-O-methylellagic acid 4'-O-xylopyranoside, and 3,4,3'-tri-O-methylflavellagic acid 4'-O-glucopyranoside.

A new hybrid pattern recognition scheme for automatic discrimination of power quality disturbances

In this paper an intelligent method for automatic detection of Power Quality Disturbances (PQDs) is presented. The proposed automatic scheme is noticed to retaining informative features or eliminating redundant features simultaneously. This paper presents an effective method, for extracting features, so-called “integrated approach”, using integration of Discrete Wavelet Transform and Hyperbolic S Transform. Moreover, a new efficient feature selection method namely Orthogonal Forward Selection by incorporating Gram Schmidt procedure and forward selection is applied for selection of the best subset features. Some different classifiers are empirically compared in order to determine the best classifier. In this automatic scheme, the variable parameters of classifiers are optimized using a powerful method namely Particle Swarm Optimization. The sensitivity of the proposed method under noisy conditions has been investigated. The average rate of correct classification using the proposed hybrid scheme for automatic discrimination of PQDs in various noise conditions 99.55% is obtained.