Bidirectional Associative Memory Research Articles

Finite-time synchronization for a fully complex-valued BAM inertial neural network with proportional delays via non-reduced order and non-separation approach

This paper focuses on finite-time synchronization (FinTS) for fully complex-valued bidirectional associative memory inertial neural networks (FCV-BAM-INNs) with proportional delays via non-reduced order and non-separation approach. Firstly, a model of the FCV-BAM-INNs with proportional delays is established. Secondly, by using the non-reduced order transformation method of inertial neural networks (INNs) and the non-separation method of complex-valued neural networks (CVNNs), an adjustable complex feedback control law with memory is designed to reach the FinTS of the drive–response FCV-BAM-INNs with proportional delays. Thirdly, some new delay-independent algebraic sufficient criteria are formed to guarantee FinTS of FCV-BAM-INNs with proportional delays by applying non smooth theory and inequalities in the complex domain, which are flexible and adjustable and easy to implement in applications. Moreover, the settling time of FinTS theoretically is calculated. Finally, some numerical simulations and an application on image encryption and decryption as evidences of the validity and availability of the results.

Finite-time ontractive stability for Fractional-Order Nonlinear Systems with delayed impulses: Applications to neural networks

This paper deals with finite-time stability (FTS) and finite-time contractive stability (FTCS) criteria of fractional-order nonlinear systems (FONSs), where the state of the systems consists of a finite number of delayed impulsive instants. To achieve these results, we have extended the theories on FTS and FTCS criteria from integer-order NSs to fractional-order systems. Employing the fractional-order Lyapunov stability theory, the sufficient conditions of the above-mentioned stability criteria are derived for a class of FONSs with state-dependent delayed impulses. Then, to ensure the applicability of the proposed results, we have analyzed the desired performances for several neural network (NN) models, such as non-autonomous NNs (NANNs), Cohen-Grossberg NNs (CGNNs), switched NNs (SNNs) and Bidirectional associative memory NNs (BAMNNs). Finally, four representative numerical examples are illustrated to demonstrate the assurance of the obtained stability conditions of the respective state-dependent impulsive fractional-order NNs (FONNs).

Hetero-associative Memory Based New Iraqi License Plate Recognition

نتيجة للتطورات الأخيرة في أبحاث الطرق السريعة بالإضافة إلى زيادة استخدام المركبات، كان هناك اهتمام كبير بنظام النقل الذكي الأكثر حداثة وفعالية ودقة (ITS) في مجال رؤية الكمبيوتر أو معالجة الصور الرقمية، يلعب تحديد كائنات معينة في صورة دورًا مهمًا في إنشاء صورة شاملة. هناك تحدٍ مرتبط بالتعرف على لوحة ترخيص السيارة (VLPR) بسبب الاختلاف في وجهة النظر، والتنسيقات المتعددة، وظروف الإضاءة غير الموحدة في وقت الحصول على الصورة والشكل واللون، بالإضافة إلى الصعوبات مثل ضعف دقة الصورة ، الصورة الباهتة ، الإضاءة السيئة، التباين المنخفض، يجب التغلب عليها. اقترحت هذه الورقة نموذجًا باستخدام تعديل الذاكرة الترابطية ثنائية الاتجاه (MBAM)، وهي نوع واحد من الذاكرة الترابطية غير المتجانسة، وتعمل MBAM على مرحلتين)مرحلتي التعلم والتقارب) للتعرف على اللوحة، ويمكن لهذا النموذج المقترح التغلب على تلك الصعوبات بسبب قدرة الذاكرة الترابطية لـ MBAM على قبول الضوضاء وتمييز الصور المشوهة، وكذلك سرعة عملية الحساب نظرًا لصغر حجم الشبكة. نتيجة دقة تحديد منطقة اللوحة هي 99.6٪، ودقة تجزئة الأحرف 98٪، والدقة المحققة للتعرف على الأحرف هي100 ٪ في ظروف مختلفة.

Non-fragile output-feedback control for delayed memristive bidirectional associative memory neural networks against actuator failure

This article investigates the stabilization property for the modeled memristive bidirectional associative memory neural networks with time-varying delay when the faulty signals received from the fluctuated controller. The non-fragile output-feedback controller is taken into account to counteract the impact of gain perturbations to end up with robust fault-tolerant setup. To tackle the weak signals in the actuator received from the fluctuated controller, control gain matrices encompass situations intended to memory non-fragile output-feedback controller. Based on the Lyapunov stability theory, differential inclusion theory, and congruence transformation, the sufficient condition for the global asymptotic stabilization property for the designed fault-tolerant memristive bidirectional associative memory neural network model is obtained in terms of linear matrix inequality by utilizing Wirtinger's inequality. Finally, numerical examples are approached with the state performance plots of the proposed memristive bidirectional associative memory neural network model with respect to the time-domain plane, to confirm the stabilization results and it illustrates the working mechanism of the designed controller.

Delay dependent complex-valued bidirectional associative memory neural networks with stochastic and impulsive effects: an exponential stability approach

Delay dependent complex-valued bidirectional associative memory neural networks with stochastic and impulsive effects: an exponential stability approach

Fixed/Preassigned-time synchronization of quaternion-valued BAM neural networks: An event-based non-separation control method

Quaternions provide expressive power beyond real numbers, allowing neural networks to capture and process correlations and patterns in data with greater complexity. Besides, event-triggering mechanism has significant advantages in reducing redundant data transmission and control costs, since the sampling instant is determined by preset trigger conditions. Based on these fact, this article investigates the fixed-time and preassigned-time synchronization of quaternion-valued Bidirectional Associative Memory (BAM) neural networks via event-triggered control. Firstly, based on a dual-layer network structure with different number of nodes, a type of BAM neural network with generalized time-varying delays is established in the quaternion domain. Secondly, four kinds of Zeno-free dynamic event-triggered control mechanisms are designed, and the trigger condition can be dynamically adjusted according to the changing requirements. In the framework of non-separation analysis, the synchronization the controlled network can be guaranteed within a fixed time, and the upper bound of the setting time is provided explicitly. Furthermore, to better meet practical needs, the above event-triggered mechanism and the criteria are extended to the PAT synchronization. Two illustrate examples are presented at last to validate the developed event-triggered mechanism and synchronization results.

Effect of Leakage Delays on Bifurcation in Fractional-Order Bidirectional Associative Memory Neural Networks with Five Neurons and Discrete Delays

Effect of Leakage Delays on Bifurcation in Fractional-Order Bidirectional Associative Memory Neural Networks with Five Neurons and Discrete Delays

Recognition of Hijaiyah Letter Patterns Using The Bidirectional Associative Memory Method

Artificial Intelligence has a very broad scope so it can solve very complex problems. Hijaiyah letter pattern recognition using the Bidirectional Associative Memory (BAM) method has the ability to detect maximum results for the blind who are learning to recognize hijaiyah letters in producing information using a bipolar system. The problem in this research is the difficulty for blind people in recognizing hijaiyah letters and patterns, so this research was designed to help blind people learn hijaiyah letters so that they can easily understand the patterns and types of letters using the 3x3 order matrix concept. In the BAM has the advantage of layers that are interconnected between one layer and another layer so that they can be connected from the input layer to the output layer. This research will produce a Hijaiyah letter recognition pattern using the 3x3 Matrix system to design a system for identifying Hijaiyah letter patterns. From the results of the tests that have been carried out, there are 5 Hijaiyah letter patterns, namely ا, ث, ج, ذ and ش which are processed by pattern recognition with a value of 1. [1.1] [1,-13],[-1,1][3,-11 ],[1,-1] [-9,-17],] [-1,1][ [13,9], [-1,1][-13,13]. Of the 5 Hijaiyah letter patterns, only 2 letter patterns were able to be recognized by the system, namely the letters ذ and ش, and 3 patterns were not able to be recognized by the system. Thus it can be concluded that not all patterns using the BAM method are able to recognize patterns precisely and correctly.

Secure adaptive event-triggered anti-synchronization for BAM neural networks with energy-limited DoS attacks

This article focuses on the problem of adaptive event-triggered anti-synchronization control for bidirectional associative memory neural networks subject to energy-limited denial of service attacks. First, a novel adaptive event-triggered scheme is developed by resorting to the acknowledgment character technique, which can help conserve valuable communication resources and has better performance in resisting malicious cyber attacks compared to traditional schemes. Second, a more general attack strategy for denial of service attacks is proposed with the consideration of energy constraints, and an anti-synchronization error system is established to analyze the anti-synchronization behavior. Then, sufficient conditions are provided to guarantee the anti-synchronization of drive and response bidirectional associative memory neural networks in H∞ sense. Next, a design approach is obtained based on the above conditions for the controller gains. Finally, a numerical example is employed to demonstrate the effectiveness of the proposed method and its superiority over the traditional event-triggered scheme.

Non-separation Method-Based Global Stability Criteria for Takagi–Sugeno Fuzzy Quaternion-Valued BAM Delayed Neural Networks Using Quaternion-valued Auxiliary Function-Based Integral Inequality

This paper focuses on the global asymptotic stability (GAS) problem for Takagi–Sugeno (T-S) fuzzy quaternion-valued bidirectional associative memory neural networks (QVBAMNNs) with discrete, distributed and leakage delays by using non-separation method. By applying T-S fuzzy model, we first consider a general form of T-S fuzzy QVBAMNNs with time delays. Then, by constructing appropriate Lyapunov–Krasovskii functionals and employing quaternion-valued integral inequalities and homeomorphism theory, several delay-dependent sufficient conditions are obtained to guarantee the existence and GAS of the considered neural networks (NNs). In addition, these theoretical results are presented in the form of quaternion-valued linear matrix inequalities (LMIs), which can be verified numerically using the effective YALMIP toolbox in MATLAB. Finally, two numerical illustrations are presented along with their simulations to demonstrate the validity of the theoretical analysis.

New Insights on Bidirectional Associative Memory Neural Networks with Leakage Delay Components and Time-Varying Delays Using Sampled-Data Control

The sampling data control of bidirectional associative memory (BAM) neural network with leakage delay is considered in this article. The BAM model is viewed as a mixed delay that combines a distributed delay, a discrete delay that varies over time, and a delay in the leaking period. The sampling system is then converted to a continuous time-delay system using an input delay method. In order to get adequate conditions in the form of linear matrix inequalities(LMIs), we build a new Lyapunov-Krasovskii Functional (LKF) in conjunction with the free weight matrix approach. Finally, a simulation results are given to show the efficiency of the theoretical approach.

Stability and Hopf Bifurcation Analysis of A Fractional-Order BAM Neural Network with Two Delays Under Hybrid Control

In this paper, considering that fractional-order calculus can more accurately describe memory and genetic properties, we introduce fractional integral operators into neural networks and discuss the stability and Hopf bifurcation of a fractional-order bidirectional associate memory (BAM) neural network with two delays. In addition, the hybrid controller is proposed to achieve Hopf bifurcation control of the system. By taking two time delays as the bifurcation parameters and analyzing of the corresponding characteristic equation, stability switching curves of the controllable system for two delays are obtained. The direction of the characteristic root crossing the imaginary axis in stability switching curves is determined. Sufficient criteria are sequentially given to judge the local stability and the existence of Hopf bifurcation of a fractional-order BAM neural network system. The numerical simulation results show that the hybrid controller can effectively control Hopf bifurcation of a fractional-order BAM neural network system with two delays.

Facial recognition based on enhanced neural network

Accurate automatic face recognition (FR) has only become a practical goal of biometrics research in recent years. Detection and recognition are the primary steps for identifying faces in this research, and The Viola-Jones algorithm implements to discover faces in images. This paper presents a neural network solution called modify bidirectional associative memory (MBAM). The basic idea is to recognize the image of a human's face, extract the face image, enter it into the MBAM, and identify it. The output ID for the face image from the network should be similar to the ID for the image entered previously in the training phase. The tests have conducted using the suggested model using 100 images. Results show that FR accuracy is 100% for all images used, and the accuracy after adding noise is the proportions that differ between the images used according to the noise ratio. Recognition results for the mobile camera images were more satisfactory than those for the Face94 dataset.

Global exponential stability of quaternion bidirectional associative memory neural networks with multiple delays

In this paper, the global exponential stability of quaternion BAM neural networks with mixed delays is investigated. The involved delays include multiple time‐varying transmission delays and unbounded distributed delays. Existence and uniqueness of equilibrium point of the considered neural networks are investigated by using the homeomorphism theory between topology spaces. To show sufficient conditions for global exponential stability of the unique equilibrium point, a method based on estimation of solutions is presented. The acquired stability conditions contain only several simple linear scalar inequalities, which can be easily solved. The availability of theoretical results is shown by a numerical example. A point worth emphasizing is that the method based on estimation of solutions is applicable for models of many time‐delay systems after a minor modification.

Bifurcation Analysis of Time-Delayed Non-Commensurate Caputo Fractional Bi-Directional Associative Memory Neural Networks Composed of Three Neurons

We are concerned in this paper with the stability and bifurcation problems for three-neuron-based bi-directional associative memory neural networks that are involved with time delays in transmission terms and possess Caputo fractional derivatives of non-commensurate orders. For the fractional bi-directional associative memory neural networks that are dealt with in this paper, we view the time delays as the bifurcation parameters. Via a standard contraction mapping argument, we establish the existence and uniqueness of the state trajectories of the investigated fractional bi-directional associative memory neural networks. By utilizing the idea and technique of linearization, we analyze the influence of time delays on the dynamical behavior of the investigated neural networks, as well as establish and prove several stability/bifurcation criteria for the neural networks dealt with in this paper. According to each of our established criteria, the equilibrium states of the investigated fractional bi-directional associative memory neural networks are asymptotically stable when some of the time delays are less than strictly specific positive constants, i.e., when the thresholds or the bifurcation points undergo Hopf bifurcation in the concerned networks at the aforementioned threshold constants. In the meantime, we provide several illustrative examples to numerically and visually validate our stability and bifurcation results. Our stability and bifurcation theoretical results in this paper yield some insights into the cause mechanism of the bifurcation phenomena for some other complex phenomena, and this is extremely helpful for the design of feedback control to attenuate or even to remove such complex phenomena in the dynamics of fractional bi-directional associative memory neural networks with time delays.

Exploration on bifurcation for an incommensurate five-neuron fractional-order BAM neural network involving multiple delays

Some qualitative analyses of bifurcations for an incommensurate fractional-order neural network with delays are considered. At first, we display an incommensurate five-neuron fractional-order bidirectional associative memory neural network (FOBAMNN) with four delays. Next, the existence and uniqueness of the solution are analyzed. Afterward, from the perspective of the two different cases of delays inducing bifurcations, some sufficient criteria of the stability and the occurrence of Hopf bifurcation are established. Some numerical simulations are designed to verify the feasibility and validity of the obtained theorems. In addition, the effects of the delays on the stability and bifurcation of the integer-order neural network and the fractional-order neural network are discussed and some meaningful results are presented in the end.

Statistical mechanics of learning via reverberation in bidirectional associative memories

We study bi-directional associative neural networks that, exposed to noisy examples of an extensive number of random archetypes, learn the latter (with or without the presence of a teacher) when the supplied information is enough: in this setting, learning is heteroassociative – involving couples of patterns – and it is achieved by reverberating the information depicted from the examples through the layers of the network. By adapting Guerra’s interpolation technique, we provide a full statistical mechanical picture of supervised and unsupervised learning processes (at the replica symmetric level of description) obtaining analytically phase diagrams, thresholds for learning, a picture of the ground-state in plain agreement with Monte Carlo simulations and signal-to-noise outcomes. In the large dataset limit, the Kosko storage prescription as well as its statistical mechanical picture provided by Kurchan, Peliti, and Saber in the eighties is fully recovered. Computational advantages in dealing with information reverberation, rather than storage, are discussed for natural test cases. In particular, we show how this network admits an integral representation in terms of two coupled restricted Boltzmann machines, whose hidden layers are entirely built of by grand-mother neurons, to prove that by coupling solely these grand-mother neurons we can correlate the patterns they are related to: it is thus possible to recover Pavlov’s Classical Conditioning by adding just one synapse among the correct grand-mother neurons (hence saving an extensive number of these links for further information storage w.r.t. the classical auto-associative setting). Likewise, it is enough to equip the BAM’s synaptic coupling with its transpose matrix (and alternate among these two, e.g., by introducing a clock) to let these networks be able count sequences of patterns (still preserving Detailed Balance, at difference w.r.t. the auto-associative case).

Delay-independent control for synchronization of memristor-based BAM neural networks with parameter perturbation and strong mismatch via finite-time technology

This paper mainly studies the synchronization problem of memristor-based bidirectional associative memory neural networks (MBAMNNs) via finite-time technology. Different from the existing neural network dynamic models, the given model in this paper is focused on the impact of parameter perturbation and strong mismatch, where strong mismatch includes parameter mismatch and time-varying delay mismatch. These characteristics can make the model be closer to the actual situation. A delay-independent feedback control scheme, which can stabilize the error system within finite-time regardless of whether the past state is known or not, is designed. It is worth noting that the constant is replaced by a function with the exponential term in the delay-independent controller, which can save the control cost to a certain extent. Based on the integral inequality technique, some sufficient conditions for MBAMNNs to converge to the equilibrium point within finite-time are provided. The validity and correctness of the theoretical results are finally confirmed by numerical simulation.

Finite-time synchronization of neutral type fractional-order bidirectional associative memory neural networks with delays via adaptive feedback control strategy

Finite-time synchronization of neutral type fractional-order bidirectional associative memory neural networks with delays via adaptive feedback control strategy

New results on finite-time projective synchronization for memristor-based hybrid delayed BAM neural networks with applications to DNA image encryption

<abstract> <p>With the popularization of digital image technology, image information has inevitably developed to involved the disclosure of personal privacy; in this study, a color image encryption algorithm was designed to encrypt and decrypt images by using chaotic sequences of a class of memristor-based hybrid delayed bidirectional associative memory neural networks (MHDBAMNNs) to protect images from illegal acquisition and use. Additionally, the discontinuity problem of the right-hand side of the Filippov system due to the hopping property of the memristor has been treated by using differential inclusion and set-valued mapping theories, and a sufficient criterion for guaranteeing the synchronization of finite-time projections derived based on the drive-response concept, Lyppunov stability theorem, and inequality technique. To improve the security performance, a color image encryption algorithm based on a combination of Chen's hyperchaotic system and a DNA codec operation was adopted, also, the robustness and validity of our proposed approach was demonstrated through image performance analysis. Furthermore, the potential application of the model in secure transmission has been explored.</p> </abstract>