Fuzzy Associative Memory Research Articles

퍼지 추론에 의한 비디오 데이터의 샷 경계 추출

본 논문에서는 퍼지 추론 방법을 이용하여 비디오 데이터에서 샷(shot)의 경계를 검출하는 방법을 제안한다. 제안된 방법에서는 컷(cut), 페이드(fade), 디졸브(dissolve)와 같은 샷의 경계들을 검출하고, 이들을 그 종류별로 분류하기 위해 기본적으로 퍼지 연상 기억장치를 확장한 퍼지 추론 방법을 이용한다. 본 논문에서는 연속적인 두 영상 사이의 차이를 나타내는 여러 특징들을 입력 퍼지 집합으로 사용하고, 샷 경계들을 출력 퍼지 집합으로 사용한다. 본 논문의 퍼지 추론 시스템은 크게 학습 단계와 추론 단계의 두 단계로 구성된다. 학습 단계에서는 퍼지 소속 함수의 결정을 통해 시스템의 기본 구조를 초기화하고 이를 바탕으로 퍼지 연상 기억장치의 학습 기능을 이용하여 퍼지 규칙을 조건부와 결론부를 연결하는 가중치의 형태로 생성한다. 그리고 추론 단계에서는 구성된 퍼지 추론 모델을 이용하여 실제 추론을 수행한다. 실험에서는 제안된 샷 경계 검출 방법의 성능을 확인하기 위해서 뉴스, 영화, 광고, 다큐멘터리, 뮤직 비디오 등의 비디오 데이터들을 활용하였다. In this paper, we describe a fuzzy inference approach for detecting and classifying shot transitions in video sequences. Our approach basically extends FAM (Fuzzy Associative Memory) to detect and classify shot transitions, including cuts, fades and dissolves. We consider a set of feature values that characterize differences between two consecutive frames as input fuzzy sets, and the types of shot transitions as output fuzzy sets. The inference system proposed in this paper is mainly composed of a learning phase and an inferring phase. In the learning phase, the system initializes its basic structure by determining fuzzy membership functions and constructs fuzzy rules. In the inferring phase, the system conducts actual inference using the constructed fuzzy rules. In order to verify the performance of the proposed shot transition detection method experiments have been carried out with a video database that includes news, movies, advertisements, documentaries and music videos.

Inference of recommendation information on the internet using improved FAM

This paper proposes a collaborative filtering system using Improved Fuzzy Associative Memory (IFAM) which re-adjusts the connection weights between the nodes of FAM using error back propagation and simplifies the fuzzy rules. The proposed technique automatically recommends high-quality information to users with similar interests on arbitrarily narrow information domains. It asks a user to rate a gauge set of items. It then evaluates the user’s rates and suggests a recommendation set of items. The proposed system is implemented in a web server and tested its performance in the domain of retrieval of technical papers, especially in the field of information technologies. The experimental results show that it may provide reliable recommendations.

Predicting Job Performance with a Fuzzy Rule-Based System

Classification problems affect all organizations. Important decisions affecting an organization's effectiveness include predicting the success of job applicants and the matching and assignment of individuals from a pool of applicants to available positions. In these situations, linear mathematical models are employed to optimize the allocation of an organization's human resources.Use of linear techniques may be problematic, however, when relationships between predictor and criterion are nonlinear. As an alternative, we developed a fuzzy associative memory (FAM: a rule-based system based on fuzzy sets and logic) and used it to derive predictive (classification) equations composed of measures of job experience and job performance. The data consisted of two job experience factors used to predict measures of job performance for four US Air Force job families. The results indicated a nonlinear relationship between experience and performance for three of the four data sets. The overall classification accuracy was similar for the two systems, although the FAM provided better classification for two of the jobs. We discuss the apparent nonlinear relationships between experience and performance, and the advantages and implications of using these systems to develop and describe behavioral models.

Maneuvering of Surface Vessels Using a Fuzzy Logic Controller

A fuzzy logic controller is developed for maneuvering control of surface vessels. The fuzzy controller uses a vessel's heading, yaw rate, distance from a reference point, and the velocity of the vessel relative to the reference point as inputs to generate the control outputs. The control outputs include rudder angle, increase in propeller thrust, and lateral bow thrust. The design of the fuzzy controller is simple and does not require a mathematical modeling of the complicated nonlinear system. The core of the fuzzy controller is a set of fuzzy associative memory (FAM) rules that correlate each group of fuzzy input sets to a fuzzy output set. A FAM rule is a logical if-then type statement based on one's sense of realism, experience, and expert knowledge. The effectiveness and robustness of the fuzzy controller are demonstrated through the numerical time-domain simulations of path tracking and dynamic positioning of a Mariner-class hull with use of nonlinear maneuvering equations of motions.

Enhanced power system stabilizer via integrated tabu–fuzzy knowledge based controller

An integrated tabu–fuzzy knowledge based controller applied to enhance the performance of power system stabilizer (PSS) is presented in this paper. The fuzzy knowledge based controller (FKBC) has been developed to perform the function of a PSS and to provide a supplementary signal to the excitation system of the synchronous generator. The method used for storing and representing the fuzzy rules is called the fuzzy associative memory (FAM) matrix. The well-defined FAM determines the performance of the FKBC and hence the tabu search algorithm is proposed and applied to determine the construction and optimization of the FAM. The controller has been tested in the case of single machine infinite bus and multi-machine systems for various types of disturbance. To show the effectiveness of the proposed controller, the comparison with the conventional PSS is presented.

The stability problem for fuzzy bidirectional associative memories

The fuzzy bidirectional associative memories (FBAMs) are studied in this paper. It is shown that any FBAM based on the S– T composition is globally stable if both S and T are triangular norms or triangular conorms. When T is an upper semicontinuous triangular norm, some necessary and sufficient conditions are established in such a way that the FBAMs based on the max - T composition are globally stable, and the stable states and the equilibria of the FBAMs were studied. When T is a continuous triangular norm, it is proved that the sequence of reverberation states of the learning FBAM is nonincreasing.

Fuzzy modeling in terms of surprise

This paper presents a new approach to fuzzy modeling based on the concept of surprise. The new concept is related to the traditional membership function by an antitone transformation. Advantages of the surprise approach include: 1. It has a consistent semantic interpretation. 2. It allows the joint use of quantitative and qualitative knowledge, using simple rules of logic. 3. It is a direct extension of (and allows combination with) the least-squares approach to reconciling conflicting approximate numerical data. 4. It is ideally suited to optimization under imprecise or conflicting goals, specified by a combination of soft and hard interval constraints. 5. It gives a straightforward approach to constructing families of functions consistent with fuzzy associative memories as used in fuzzy control, with tuning parameters (reflecting linguistic ambiguity) that can be adapted to available performance data.

Forecasting of engineering manpower through fuzzy associative memory neural network with ARIMA: a comparative study

The smooth working of industry depends on the availability of proper engineering manpower. If proper qualified and experienced technical personnel are not available, the industry cannot run in the most efficient way. Here, an effort is made to assess the engineering manpower requirement (personnel belonging to mechanical engineering) in certain industry group (steel manufacturing) in the state of West Bengal in India for the next 5 years. The method of auto regressive integrated moving average (ARIMA) and the fuzzy associative memory (FAM) neural network model are tested and based on error analysis (calculation of average error) the model with minimum error is selected and used for assessment of futuristic engineering manpower. Certain statistical functions, i.e. regression analysis using a least square technique based on linear, exponential, curvilinear (parabolic) equations and the tables of Orthogonal Polynomial are applied on the estimated data value calculated earlier. The particular statistical model is chosen based on the average error of estimated date generated using statistical models with the actual data over span of years. The said statistical model based on the estimated data using the selected model of ARIMA or FAM neural network can be used for the generation of futuristic forecasted engineering manpower.

A neural-fuzzy classifier for recognition of power quality disturbances

This paper presents a neural-fuzzy technology-based classifier for the recognition of power quality disturbances. The classifier adopts neural networks in the architecture of frequency sensitive competitive learning and learning vector quantization (LVQ). With given size of codewords, the neural networks are trained to determine the optimal decision boundaries separating different categories of disturbances. To cope with the uncertainties in the involved pattern recognition, the neural network outputs, instead of being taken as the final classification, are used to activate the fuzzy-associative-memory (FAM) recalling for identifying the most possible type that the input waveform may belong to. Furthermore, the input waveforms are preprocessed by the wavelet transform for feature extraction so as to improve the classifier with respect to recognition accuracy and scheme simplicity. Each subband of the transform coefficients is then utilized to recognize the associated disturbances.

A Web-based collaborative filtering system

In this paper we describe a collaborative filtering system for automatically recommending high-quality information to users with similar interests on arbitrarily narrow information domains. It asks a user to rate a gauge set of items. It then evaluates the user's rates and suggests a recommendation set of items. We interpret the process of evaluation as an inference mechanism that maps a gauge set to a recommendation set. We accomplish the mapping with fuzzy associative memory. We implemented the suggested system in a Web server and tested its performance in the domain of retrieval of technical papers, especially in the field of information technologies. The experimental results show that it may provide reliable recommendations.

Dynamic positioning of drilling vessels with a fuzzy logic controller

A fuzzy logic controller for dynamic positioning of drilling vessels in deep water is presented. The core of the fuzzy controller is a set of fuzzy associative memory (FAM) rules that correlate each group of fuzzy control input sets to a fuzzy control output set. A FAM rule is a logical if-then-type statement based on one's sense of realism and experience or can be provided by an expert operator. The design of the fuzzy controller is very simple and does not require mathematical modelling of the complicated nonlinear system based on first principles. The fuzzy controller uses measured vessel heading, yaw rate, distance and velocity of the vessel relative to the desired position (location and heading) to generate the control outputs to bring the vessel to and maintain it in the desired position. The control outputs include the rudder angle, propeller thrust and lateral bow thrust. The effectiveness and robustness of the fuzzy controller are demonstrated through numerical time-domain simulations of the dynamic positioning of a drill ship of Mariner Class hull with use of nonlinear ship equations of motions.

Fractional correlation filter for fuzzy associative memories

The optical fractional correlation is a very useful tool for partially controllable spatially varying correlation-based pattern recognition. It is shown to be a good match to fuzzy associative memory that performs partial fuzzy associations in parallel. We exploit the potential application of the spatially varying fractional correlation in implementing parallel fuzzy association. Computer simulations illustrate our conclusions.

Fuzzy associative memories for the computation of the bandwidth of rectangular microstrip antennas with thin and thick substrates

A method based on fuzzy associative memories for calculating the bandwidth of electrically thin and thick rectangular microstrip patch antennas is presented. The design parameters of the fuzzy associative memories are optimally determined by using the classical, modified and improved tabu search algorithms. The theoretical bandwidth results obtained by using the fuzzy associative memories are in very good agreement with the experimental results available in the literature.

Fuzzy logical bidirectional associative memory

We introduce a bidirectional associative memory (BAM). The stable points of the memory are naturally interpreted as (non-sharp) concepts – the memory performs association of extents and intents of concepts. We show that this memory is stable and that the set of all stable points forms a complete lattice. We propose a learning algorithm and prove that it enables perfect learning provided the training set forms a consistent conceptual structure. Examples demonstrating the results are presented. Unlike in the case of other associative memories (M. Arbib (Ed.), The Handbook of Brain Theory and Neural Networks, MIT Press, London, 1995) the formal apparatus, architecture, dynamics and convergence proof etc. are based on algebraic structures of fuzzy logic in narrow sense.

Implementation of an Intelligent Control System Using Fuzzy ITI

This paper proposes a new multi-strategy (hybrid) intelligent control technique whose concept is applicable to the control of a wide range of processes The proposed technique uses Incremental Tree Induction (ITI) as the learning algorithm, and incor porates fuzzy logic to deal with uncertainties apparent in the process to be controlled. ITI operates solely on symbolic fuzzy knowledge, as both the input features (data obtained from the process to be controlled) and the output decisions of the intelligent controller are described by fuzzy linguistic variables. Fuzzy associative memories (FAMs) are employed to store and manage the fuzzy knowledge, and are simulated operationally by fuzzy binary decision trees which encode the input-output space. The main novelty of this approach is the automatic synthesis (both off-line and on-line) of multi-dimensional FAMs from inception, in contrast to the manual implementation of traditional FAMs. A second important advantage is the self-explanatory nature of the FAMs (and their underlying control laws) generated by this approach. The new technique is demonstrated in its application to the intelligent navigation of a mobile robot.

Plastic Manufacturing Process Selection Methodology Using Self-Organising Map (SOM)/Fuzzy Analysis

For plastic products, one of the crucial decisions made during the preliminary design stage is the selection of an appropriate manufacturing process. In many industries, the selection of the manufacturing process is primarily based on the empirical knowledge and past experience of the manufacturing personnel. This selection procedure may result in inconsistent or poor choices if the manufacturing personnel fail to map correctly the product characteristics with the manufacturing efficacy of various manufacturing processes. This paper presents an intelligent self-organising map (SOM)/fuzzy-based model to aid designers in the selection of an appropriate plastic manufacturing process. The plastic part attributes are broadly classified into three main categories: part characteristics, material type, and production requirements. A fuzzy membership function is generated for each of the attributes using the self-organising map paradigm. Fuzzy associative memories (FAMs) are used to perform reasoning on these input fuzzy sets to derive the output fuzzy sets. The output fuzzy sets are then defuzzified to determine the process compatibility scores (PCS). The working of the proposed model is demonstrated using an example case.

Neural fuzzy relational systems with a new learning algorithm

Fuzzy relational systems can represent symbolic knowledge in a formal numerical (subsymbolic) framework, with the aid of fuzzy relation equations. The disadvantage of this methodology is the need for a priori knowledge in order to construct the fuzzy relation equation. In this paper, a neural network model is proposed in order to represent fuzzy relational systems without the need of the construction of the fuzzy relation equation. The network ensures the ideal perfect recall of fuzzy associative memories when the a posteriori constructed fuzzy relation equation has a non-empty solution set. It is actually a single layer of generalized neurons (compositional neurons) that perform the sup- t-norm composition. An on-line learning algorithm adapting the weights of its interconnections is incorporated into the neural network. These weights are actually the elements of the fuzzy relation representing the fuzzy relational system. The algorithm is based on the knowledge about the topographic structure of the respective fuzzy relation.

A new method for extracting the rules of the Fuzzy Associative Memory

Based on the features of the Fuzzy Associative Memory (FAM), a method for extracting the FAM rules is proposed. When the partition of the input space is available, the extraction of the FAM rules can be converted into the plane clustering. When the number of the linguistic values of the output variable is given, the steady-state rule can be first obtained with the one-dimensional frequency histograms of the input and output variables, and then the extraction of the remaining rules can be converted into the partition of the input space with the tree classifier through clustering in the input-output product space the planes vertical to the output axis. Finally the property of the method is tested with an example of the truck backer-upper control system.

The fuzzy associative memory of max-min fuzzy neural network with threshold

In this paper, we introduce the max-min fuzzy neural network with threshold which generalizes the fuzzy neural network models in [1,4–;6,8,10] and show that the storage capacity of the two-layer max-min fuzzy neural network is identical with one of the max-min fuzzy neural network with hidden layers. For given fuzzy pattern pairs ( X 1, Y 1),…, ( X p , Y p ), we obtain an equivalent condition which the family of given fuzzy pattern pairs can be stored in the fuzzy neural network. Finally, we give a example to demonstrate our conclusions.

Optical triangular-partition fuzzy systems with one-memory-matrix fuzzy associative memory

Taking advantage of the special properties of fuzzy sets, we propose that one-memory-matrix fuzzy associative memory can be used to realize triangular-partition fuzzy inference systems efficiently. The proposed model has perfect recall capability.