Dynamic Pattern Recognition Research Articles

VYPER: Vulnerability detection in binary code

AbstractThis paper presents a method for exploitable vulnerabilities detection in binary code with almost no false positives. It is based on the concolic (a mix of concrete and symbolic) execution of software binary code and the annotation of sensitive memory zones of the corresponding program traces (represented in a formal manner). Three big families of vulnerabilities are considered (taint related, stack overflow, and heap overflow). Based on the angr framework as a supporting software VulnerabilitY detection based on dynamic behavioral PattErn Recognition was developed to demonstrate the viability of the method. Several test cases using custom code, Juliet test base and widely used public libraries were performed showing a high detection potential for exploitable vulnerabilities with a very low rate of false positives.

Intelligent Control of Flexible Joint Manipulator Based on Dynamic Pattern Recognition

With the rapid development of intelligent manufacturing and Internet technology, the industrial system has entered a new stage of development. As an indispensable carrier for intelligent manufacturing and industrial development, robots are expanding their applications. Among them, the flexible mechanical arm has the advantages of light weight, low energy consumption and low inertia compared with the bulky rigid mechanical arm, and has been increasingly valued. The flexible manipulator is a very complex dynamic system whose dynamic equations are characterized by nonlinearity, strong coupling and time-varying. Therefore, this paper uses the most common and effective method to establish the dynamic model of the flexible manipulator using the Lagrange equation. Due to the uncertain system parameters, lack of control of the trajectory and the influence of load changes and external disturbances, the flexible manipulator has great uncertainty in its control process, and the traditional control methods have not very good control effect. Based on this, this paper proposes a combination of dynamic pattern recognition theory and flexible joint manipulator intelligent control method for the two-link flexible manipulator, and uses the new GA-RBF neural network closed-loop adaptive control method to achieve high precision. Trajectory tracking ensures stability in a shorter time. The simulation results show that the intelligent joint control method based on dynamic pattern recognition has better trajectory tracking and autonomous fast recognition dynamic mode.

Abrupt stall detection for axial compressors with non-uniform inflow via deterministic learning

This paper focuses on the early detection problem of abrupt stall inception in axial compressors subject to non-uniform inflow. The considered stall inception is small and abrupt. Such a stall inception is not only easy to cause system instability, but also difficult to be rapidly detected. In this paper, a novel abrupt stall detection approach is proposed for axial compressors with non-uniform inflow by using the deterministic learning algorithm. Firstly, a high-order distortion dynamic mode is derived to generate the normal and stall inception situations. Subsequently, dynamic features of these different situations are extracted and stored in constant radial basis function networks, which are regarded as the training patterns. Based on dynamic pattern recognition, the average L1 norms of the state errors are used as the measure of the dynamical differences between the test pattern and the training patterns, and then a stall detection method is developed to rapidly decide whether the test patterns are the abrupt stall inception. Simulation studies are given to show that the proposed stall detection approach is sensitive to small dynamic changes of axial compressors, thereby rapidly detecting the abrupt stall inception.

A Robust and Scalable Method for Infrared Target Identification

Recently, real-time and robust infrared target identification methodshave become critical for many applications in almost all kinds of modern weapon systems, including infrared search and tracking systems,surveillance systems and so on. However, at the long distance noises contaminated infrared imaging which the targets only occupied several pixels and the lowest entropy of information to obscure recognition the targets with digital image processing approaches effectively. To address these issues, a robust and scalable method for infrared target identification was proposed. With non-local and dynamic pattern recognition, the template matching accuracy between template and target improved significantly. It demonstrated that more than 80% accurate prediction rate of the proposed method in the practicing applications.

Exploring the physiological information of Sun-induced chlorophyll fluorescence through radiative transfer model inversion

A novel approach to characterize the physiological conditions of plants from hyperspectral remote sensing data through the numerical inversion of a light version of the SCOPE model is proposed. The combined retrieval of vegetation biochemical and biophysical parameters and Sun-induced chlorophyll fluorescence (F) was investigated exploiting high resolution spectral measurements in the visible and near-infrared spectral regions. First, the retrieval scheme was evaluated against a synthetic dataset. Then, it was applied to very high resolution (sub-nanometer) canopy level spectral measurements collected over a lawn treated with different doses of a herbicide (Chlorotoluron) known to instantaneously inhibit both Photochemical and Non-Photochemical Quenching (PQ and NPQ, respectively). For the first time the full spectrum of canopy F, the fluorescence quantum yield (ΦF), as well as the main vegetation parameters that control light absorption and reabsorption, were retrieved concurrently using canopy-level high resolution apparent reflectance (ρ*) spectra. The effects of pigment content, leaf/canopy structural properties and physiology were effectively discriminated. Their combined observation over time led to the recognition of dynamic patterns of stress adaptation and stress recovery. As a reference, F values obtained with the model inversion were compared to those retrieved with state of the art Spectral Fitting Methods (SFM) and SpecFit retrieval algorithms applied on field data. ΦF retrieved from ρ* was eventually compared with an independent biophysical model of photosynthesis and fluorescence. These results foster the use of repeated hyperspectral remote sensing observations together with radiative transfer and biochemical models for plant status monitoring.

A recursive patterns matching model for the dynamic pattern recognition problem

ABSTRACTThis paper defines a new recursive pattern matching model based on the theory of the systemic functioning of the human brain, called pattern recognition theory of mind, in the context of the dynamic pattern recognition problem. Dynamic patterns are characterized by having properties that change in intervals of time, such as a pedestrian walking or a car running (the negation of a dynamic pattern is a static pattern). Novel contributions of this paper include: (1) Formally develop the concepts of dynamic and static pattern, (2) design a recursive pattern matching model, which exploits the idea of recursivity and time series in the recognition process, and the unbundling/integration of pattern to recognize, and (3) develop strategies of pattern matching from two major orientations: recognition of dynamic patterns oriented by characteristic, or oriented by perception. The model is instantiated in several cases, to analyze its performance.

Design of an Adaptive Human-Machine System Based on Dynamical Pattern Recognition of Cognitive Task-Load.

This paper developed a cognitive task-load (CTL) classification algorithm and allocation strategy to sustain the optimal operator CTL levels over time in safety-critical human-machine integrated systems. An adaptive human-machine system is designed based on a non-linear dynamic CTL classifier, which maps a set of electroencephalogram (EEG) and electrocardiogram (ECG) related features to a few CTL classes. The least-squares support vector machine (LSSVM) is used as dynamic pattern classifier. A series of electrophysiological and performance data acquisition experiments were performed on seven volunteer participants under a simulated process control task environment. The participant-specific dynamic LSSVM model is constructed to classify the instantaneous CTL into five classes at each time instant. The initial feature set, comprising 56 EEG and ECG related features, is reduced to a set of 12 salient features (including 11 EEG-related features) by using the locality preserving projection (LPP) technique. An overall correct classification rate of about 80% is achieved for the 5-class CTL classification problem. Then the predicted CTL is used to adaptively allocate the number of process control tasks between operator and computer-based controller. Simulation results showed that the overall performance of the human-machine system can be improved by using the adaptive automation strategy proposed.

Detection of tropical cyclone genesis via quantitative satellite ocean surface wind pattern and intensity analyses using decision trees

Abstract Microwave remote sensing can be used to measure ocean surface winds, which can be used to detect tropical cyclone (TC) formation in an objective and quantitative way. This study develops a new model using WindSat data and a machine learning approach. Dynamic and hydrologic indices are quantified from WindSat wind and rainfall snapshot images over 352 developing and 973 non-developing tropical disturbances from 2005 to 2009. The degree of cyclonic circulation symmetry near the system center is quantified using circular variances, and the degree of strong wind aggregation (heavy rainfall) is defined using a spatial pattern analysis program tool called FRAGSTATS. In addition, the circulation strength and convection are defined based on the areal averages of wind speed and rainfall. An objective TC formation detection model is then developed by applying those indices to a machine-learning decision tree algorithm using calibration data from 2005 to 2007. Results suggest that the circulation symmetry and intensity are the most important parameters that characterize developing tropical disturbances. Despite inherent sampling issues associated with the polar orbiting satellite, a validation from 2008 to 2009 shows that the model produced a positive detection rate of approximately 95.3% and false alarm rate of 28.5%, which is comparable with the pre-existing objective methods based on cloud-pattern recognition. This study suggests that the quantitative microwave-sensed dynamic ocean surface wind pattern and intensity recognition model provides a new method of detecting TC formation.

SVM을 이용한 교전영역 내 위협목록 획득방법

This paper presents a threat list acquisition method in an engagement area using the support vector machines (SVM). The proposed method consists of track creation, track estimation, track feature extraction, and threat list classification. To classify the threat track robustly, dynamic track estimation and pattern recognition algorithms are used. Dynamic tracks are estimated accurately by approximating a track movement using position, velocity and time. After track estimation, track features are extracted from the track information, and used to classify threat list. Experimental results showed that the threat list acquisition method in the engagement area achieved about 95 % accuracy rate for whole test tracks when using the SVM classifier. In case of improving the real-time process through further studies, it can be expected to apply the fire control systems.

Mapping Temporal Variables Into the NeuCube for Improved Pattern Recognition, Predictive Modeling, and Understanding of Stream Data.

This paper proposes a new method for an optimized mapping of temporal variables, describing a temporal stream data, into the recently proposed NeuCube spiking neural network (SNN) architecture. This optimized mapping extends the use of the NeuCube, which was initially designed for spatiotemporal brain data, to work on arbitrary stream data and to achieve a better accuracy of temporal pattern recognition, a better and earlier event prediction, and a better understanding of complex temporal stream data through visualization of the NeuCube connectivity. The effect of the new mapping is demonstrated on three benchmark problems. The first one is the early prediction of patient sleep stage event from temporal physiological data. The second one is the pattern recognition of dynamic temporal patterns of traffic in the Bay Area of California and the last one is the Challenge 2012 contest data set. In all the cases, the use of the proposed mapping leads to an improved accuracy of pattern recognition and event prediction and a better understanding of the data when compared with traditional machine learning techniques or SNN reservoirs with an arbitrary mapping of the variables.

Self-Organization of Spatio-Temporal Hierarchy via Learning of Dynamic Visual Image Patterns on Action Sequences.

It is well known that the visual cortex efficiently processes high-dimensional spatial information by using a hierarchical structure. Recently, computational models that were inspired by the spatial hierarchy of the visual cortex have shown remarkable performance in image recognition. Up to now, however, most biological and computational modeling studies have mainly focused on the spatial domain and do not discuss temporal domain processing of the visual cortex. Several studies on the visual cortex and other brain areas associated with motor control support that the brain also uses its hierarchical structure as a processing mechanism for temporal information. Based on the success of previous computational models using spatial hierarchy and temporal hierarchy observed in the brain, the current report introduces a novel neural network model for the recognition of dynamic visual image patterns based solely on the learning of exemplars. This model is characterized by the application of both spatial and temporal constraints on local neural activities, resulting in the self-organization of a spatio-temporal hierarchy necessary for the recognition of complex dynamic visual image patterns. The evaluation with the Weizmann dataset in recognition of a set of prototypical human movement patterns showed that the proposed model is significantly robust in recognizing dynamically occluded visual patterns compared to other baseline models. Furthermore, an evaluation test for the recognition of concatenated sequences of those prototypical movement patterns indicated that the model is endowed with a remarkable capability for the contextual recognition of long-range dynamic visual image patterns.

A novel approach to decision making of Mined Data using Dynamic Snapshot Pattern Recognition Algorithm (DS-PRA)

A new approach to pattern recognition and decision machines profiling is proposed, proved and tested. The technique adopts the Snapshot method dynamically as a function of both organization policy and the organization departments policies. Such policies are associated with individual products and services provided by the organization with departments policies derived from general organization profile with the organization policy being a function of the various departments profiles. It is proved through real data the ability of such algorithm to classify, detect and predict policy changes and identify differences between different organizations. Also, such algorithm combines the concepts of general Artificial intelligence through the use of knowledge bases and Neural Networks by utilizing a similar weights matrix.

Profiling Hoodia Extracts by HPLC with Charged Aerosol Detection and Electrochemical Array Detection and Pattern Recognition

Hoodigosides are oxypregnane steroidal glycosides abundant in Hoodia gordonii and related plants native to the deserts of southwestern Africa. These plants, used traditionally to ease hunger, enjoy wide use today in dietary supplements purported to suppress appetite and aid in weight loss. Presented are two approaches to profile Hoodia related products. First, Hoodia extracts were analyzed by HPLC with charged aerosol detection. Eight hoodigosides isolated from dried plant material were separated within 15 min on a solid core enhanced C8 analytical column. With low-nanogram sensitivity, the charged aerosol detector responds uniformly to all non-volatile species including the target analytes, degradation products and impurities that may not possess a chromophore. Second, Hoodia extracts were analyzed by gradient HPLC with an online electrochemical array detector that responds to redox active compounds such as polyphenols. Each of the sixteen sensors in the electrochemical array detects analytes at a unique potential to yield picogram on-column sensitivity for diverse compounds that vary widely in redox properties. The chromatograms resulting from this voltammetric approach often clearly resolve co-eluting compounds. These two techniques provide complementary data that were used to quantify known compounds. The complex patterns of both known and unknown compounds revealed by the detectors were also interrogated by pattern recognition software to support inferences on product quality, authenticity, adulteration, or origin. Limits of detection, linear and dynamic range, precision, and pattern recognition results are compared for the two approaches.

3D Global Dynamic Window Approach for Navigation of Autonomous Underwater Vehicles

An autonomous unmanned underwater vehicle is a type of marine self-propelled robot that executes some specific mission and returns to base on completion of the task. In order to successfully execute the requested operations, the vehicle must be guided by an effective navigation algorithm that enables it to avoid obstacles and follow the best path. Architectures and principles for intelligent dynamic systems are being developed, not only in the underwater arena but also in related areas where the work does not fully justify the name. The problem of increasing the capacity of systems management is highly relevant based on the development of new methods for dynamic analysis, pattern recognition, artificial intelligence, and adaptation. Among the large variety of navigation methods that presently exist, the dynamic window approach is worth noting. It was originally presented by Fox et al. and has been implemented in indoor office robots. In this paper, the dynamic window approach is applied to the marine world by developing and extending it to manipulate vehicles in 3D marine environments. This algorithm is provided to enable efficient avoidance of obstacles and attainment of targets. Experiments conducted using the algorithm in MATLAB indicate that it is an effective obstacle avoidance approach for marine vehicles.

Dynamic supervised classification method for online monitoring in non-stationary environments

The monitoring of a system functioning is achieved using a classifier which determines at each instant the class of a new incoming pattern. In non-stationary environments, the classifier must be able to adjust its parameters according to changes in the environment conditions. This requires a continuous learning while new patterns are available. Incremental learning is an efficient continuous learning technique for updating the classifier parameters without starting from scratch every time a new pattern is available. However in non-stationary environments, data characteristics may drift over time. This leads to deteriorate dramatically the performance of incremental learning algorithms over time. This is due to the use of data which is no more consistent with the characteristics of new incoming data. Thus, a mechanism to use only the recent and representative patterns to update the classifier parameters without a “catastrophic forgetting” is necessary. In this paper, we propose a dynamic pattern recognition method, named Dynamic Fuzzy Pattern Matching, to be used for the online monitoring of non-stationary processes functioning. This method is based on the use of an incremental algorithm allowing to follow the accumulated gradual changes of classes characteristics after the classification of each new pattern. When the accumulated gradual changes reach a suitable predefined threshold, the classifier parameters are adapted online using the recent and useful patterns.

SP0074 MRI in Psoriatic Arthritis - From Spine to Toes

Psoriatic arthritis (PsA) is one of the most exciting joint diseases. The clinical appearance of PsA is very diverse, involving the spine, sacroiliac joints, peripheral joints and entheses. Consequently, also...

SP0041 MRI in Psoriatic Arthritis - From Spine to Toes

Psoriatic arthritis (PsA) is one of the most exciting joint diseases. The clinical appearance of PsA is very diverse, involving the spine, sacroiliac joints, peripheral joints and entheses. Consequently, also...

Neurodynamic System Theory and Applications

Neurodynamical systems have gradually become a popular research topic owing to their broad applications in such fields as optimization, linear and nonlinear programming, associative memory, pattern recognition, and computer vision. In view of some inevitable factors, there have been formed various neurodynamical systems including delayed neural networks, stochastic neural networks, impulsive neural networks, reaction-diffusion neural networks, and fuzzy neural networks. Over the last few decades, considerable attention has been devoted to this research area not only for enriching the theory of differential equations and dynamical systems but also for deeply understanding the dynamic states of neural networks for better modelling the brain. The current special issue puts its emphasis on the study of neurodynamical system theory and applications. Call for papers has been carefully prepared by the guest editors and posted on the journal’s web page, which has received many attentions followed by some submissions among wide topics such as delayed neural systems, stochastic neural systems, impulsive neural systems, reaction-diffusion neural systems, fuzzy neural systems, evolutionary neural systems, mathematical modeling of neural systems, computational neuroscience, neurodynamical optimization and adaptive dynamic programming, cognitive models, pattern recognition, and neural network applications. All manuscripts submitted to this special issue went through a thorough peer-refereeing process. Based on the reviewers’ reports, eleven original research articles are finally accepted. The contents embrace the synchronization of coupled neural networks, the numerical analysis of stochastic delayed partial differential equations, and the stability analysis of delayed impulsive reaction-diffusion neural networks and switched neural networks. It is certainly impossible to provide in this short editorial a more comprehensive description for all articles in this special issue. However, the team of the guest editors believes that the results included reflect some recent trends in research and outline new ideas for future studies of neurodynamical system theory and applications.

Human gait recognition via deterministic learning

Recognition of temporal/dynamical patterns is among the most difficult pattern recognition tasks. Human gait recognition is a typical difficulty in the area of dynamical pattern recognition. It cla...

Human gait recognition via deterministic learning

Recognition of temporal/dynamical patterns is among the most difficult pattern recognition tasks. Human gait recognition is a typical difficulty in the area of dynamical pattern recognition. It classifies and identifies individuals by their time-varying gait signature data. Recently, a new dynamical pattern recognition method based on deterministic learning theory was presented, in which a time-varying dynamical pattern can be effectively represented in a time-invariant manner and can be rapidly recognized. In this paper, we present a new model-based approach for human gait recognition via the aforementioned method, specifically for recognizing people by gait. The approach consists of two phases: a training (learning) phase and a test (recognition) phase. In the training phase, side silhouette lower limb joint angles and angular velocities are selected as gait features. A five-link biped model for human gait locomotion is employed to demonstrate that functions containing joint angle and angular velocity state vectors characterize the gait system dynamics. Due to the quasi-periodic and symmetrical characteristics of human gait, the gait system dynamics can be simplified to be described by functions of joint angles and angular velocities of one side of the human body, thus the feature dimension is effectively reduced. Locally-accurate identification of the gait system dynamics is achieved by using radial basis function (RBF) neural networks (NNs) through deterministic learning. The obtained knowledge of the approximated gait system dynamics is stored in constant RBF networks. A gait signature is then derived from the extracted gait system dynamics along the phase portrait of joint angles versus angular velocities. A bank of estimators is constructed using constant RBF networks to represent the training gait patterns. In the test phase, by comparing the set of estimators with the test gait pattern, a set of recognition errors are generated, and the average L(1) norms of the errors are taken as the similarity measure between the dynamics of the training gait patterns and the dynamics of the test gait pattern. Therefore, the test gait pattern similar to one of the training gait patterns can be rapidly recognized according to the smallest error principle. Finally, experiments are carried out on the NLPR and UCSD gait databases to demonstrate the effectiveness of the proposed approach.