By Topic

Soft Computing, Computing with Words and Perceptions in System Analysis, Decision and Control, 2009. ICSCCW 2009. Fifth International Conference on

Date 2-4 Sept. 2009

Filter Results

Displaying Results 1 - 25 of 78
  • [Front cover]

    Publication Year: 2009 , Page(s): c1
    Save to Project icon | Request Permissions | PDF file iconPDF (91 KB)  
    Freely Available from IEEE
  • Particle swarm optimization methods for data clustering

    Publication Year: 2009 , Page(s): 1 - 6
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (424 KB) |  | HTML iconHTML  

    This paper discusses the application of particle swarm optimization (PSO) to data clustering. Four different methods of PSO are tested on six test data sets and compared to k-means and fuzzy c-means. The four PSO methods, combinations of the constriction method, inertia, and the predator-prey method all out-perform k-means and fuzzy c-means in all test cases to varying degrees in terms of quantization error. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Seamless image stitching algorithm using radiometric lens calibration for high resolution optical microscopy

    Publication Year: 2009 , Page(s): 1 - 4
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1505 KB) |  | HTML iconHTML  

    Image mosaicing method is an image-based rendering and visualization method which is a common procedure in the generation of panoramic (composite) images and applications, such as creating virtual reality, super resolution, object insertion and object removal, etc. Specifically, the image stitching approach becomes common in super resolution digital imaging microscopy, where it is required to achieve both high magnification and a large field of view. This work presents the algorithm for the stitching of different overlapped high resolution views of the whole scene obtained from high magnification optical microscopes into a large radiometric balanced image through maximization of a cross-correlation function. To seamlessly blend and reduce the possibility of intensity and color mismatching of multiple source images in mosaicing, a radiometric calibration of the microscope optical system is employed. The algorithm was tested with grayscale and color images, and showed convincing and robust results in image stitching and blending. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Diagnosis of diabetes by using adaptive neuro fuzzy inference systems

    Publication Year: 2009 , Page(s): 1 - 4
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (2568 KB) |  | HTML iconHTML  

    Most of discoveries indicate that the best way to overcome diabetes is to prevent the risks of diabetes before becoming a diabetic. With this opinion, we would like to find a way to estimate diabetes risk, according to some variables such as age, total cholesterol, gender or shape of the body. Due to having fuzzy input and output (glucose rate) values and because of that dependent variable have more than 2 values (unlike binary logic), ANFIS and Multinomial Logistic Regression should be executed for comparison. Then the results were benchmarked. As a result, in case of that there is a system which contains fuzzy inputs and output, ANFIS gives better results than Multinomial Logistic Regression for diabetes diagnosis. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • ANFIS supported question classification in computer adaptive testing (CAT)

    Publication Year: 2009 , Page(s): 1 - 4
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (2561 KB) |  | HTML iconHTML  

    E-learning has become a major trend in the computer assisted teaching with the rapid development of Internet technologies. Web-based education is a very important component of education technology. One of the main advantage is the classroom and platform independence. Implementing Artificial Intelligence (AI) techniques to support efforts to improve the Web's intelligence and provide better services to the end users. In this study, three popular AI methods: Artificial Neural Network (ANN), Support Vector Machines (SVM), and Adaptive Network Based Fuzzy Inference System (ANFIS) were benchmarked in terms of effectiveness and performance within a Web-based environment. As the pilot test, ¿History of Civilization¿ class was selected. The question classification abilities depending on the item responses of students, item difficulties of questions, and question levels were determined by using Gaussian Normal Curve. Comparison study was conducted by considering the performance and class correctness of the sample questions (n=13) by using the given three inputs. The results showed that ANFIS has better performance than ANN and AVM in web-based education. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • PMDC motor speed control with fuzzy logic algorithm using PIC16F877 micro controller and plotting data on monitor

    Publication Year: 2009 , Page(s): 1 - 4
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (891 KB) |  | HTML iconHTML  

    In this study, speed of permanent magnet dc (PMDC) motor is controlled with fuzzy logic (FL) algorithm at real time. For real time closed loop control, it has been used a PIC16F877 micro controller usage with 8 bit resolution. Only 3 membership functions which are called Negative (N), Zero (Z) and Positive (P) are used definition for FL algorithm, because of the difficulty to write the software to the micro controller via compiler program. Membership degrees of the functions are defined from 0 to 255 which is 8 bit unsigned maximum value. Integer numbers are used for programming algorithm, because compiler program using to write the software program to the micro controller is not enough sensitive for floating point number. Instantaneous values of the speed of the motor are transferred to PC with another PIC16F877 micro controller and plotting on monitor using interface program developed from DELPHI. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Weighted majority voting for face recognition from low resolution video sequences

    Publication Year: 2009 , Page(s): 1 - 4
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (534 KB) |  | HTML iconHTML  

    In this paper a new system for recognizing faces from video sequences using weighted majority voting (WMV) method is proposed. In the training phase, the system uses principle component analysis (PCA) based single eigenspace generated by sequences of faces of all subjects with the same resolution. For the testing phase, the system employs several preprocessing tasks whereby for all subjects' videos, the face images with varying resolutions in different frames are automatically extracted, histogram equalized to alleviate the effects of changing illumination, and upsampled to the resolution of the eigenfaces. For the recognition phase, each recognized subject is assigned a weight based on a measure of information capacity of each tested frame. Finally the subject with highest cumulative weight, through the video sequence is declared to be the recognized person. The proposed WMV system is robust to scale changes and effectively addresses the problem of recognition from low resolution video sequences. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An auction based mathematical model and heuristics for resource co-allocation problem in grids and clouds

    Publication Year: 2009 , Page(s): 1 - 4
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (320 KB) |  | HTML iconHTML  

    Resource co-allocation problem is one of the challenging problems in grid and cloud environments. In this paper, we propose an auction based mathematical model for providing economically efficient allocations of resources in these environments. Our model improves our earlier multi-unit nondiscriminatory combinatorial auction model by introducing a penalty cost mechanism which results in better utilization of resources and hence increasing revenue depending on the application domain. The model is formulated using integer linear programming. Five new heuristic algorithms based on the well-known meta-heuristic techniques are proposed. An intelligent neighbor selection method is also introduced. The performances of the algorithms are compared with a commercial mixed integer programming (MIP) solver on generated test cases. The solutions provided by the evolutionary algorithm are as good as the solutions provided by the MIP solver for these test cases. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fault classification in gears using support vector machines (SVMs) and signal processing

    Publication Year: 2009 , Page(s): 1 - 4
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (488 KB) |  | HTML iconHTML  

    This study presents a procedure for gear fault identification based on vibration signal processing techniques and support vector machines (SVMs). The required feature vector is extracted from vibration signals by time, frequency and time-frequency analysis. A feature selection technique based on Euclidian distance is utilized and five salient features are selected from the original feature set. These features are fed into the classification algorithm. Gear conditions considered were healthy, slightly worn, medium worn and broken-teeth gears. The output of classifier algorithm indicates the status of the gearbox by four labels. The results show that the developed SVM-based procedure is able to discriminate the faults clearly. The effectiveness of the feature selection method is demonstrated by experiments. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Constructing robot's model of external environment on basis of linguistic relations and generalized constraints

    Publication Year: 2009 , Page(s): 1 - 3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (426 KB) |  | HTML iconHTML  

    The problem of constructing internal model of external environment for intelligent robot is considered. Basic types of relations are discussed to specify robot's situational behavior. A new approach to linguistic internal model specification based on the synthesis of Pospelov's logical-linguistic modeling and Zadeh's generalized constraints is suggested. A short presentation of generalized constraints types and techniques is given. The expression of space and time relations in the form of generalized constraints is proposed; some relevant examples are shown. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Design team capability and project progress evaluation based on information monitoring technology

    Publication Year: 2009 , Page(s): 1 - 3
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (450 KB) |  | HTML iconHTML  

    Various fields of human activity (management, politics, marketing, medicine, etc.) are involved in estimating the current state and predicting the development of some complex system or process with imprecise and unreliable data. In this paper, we introduce an approach that can help deal with such tasks. We describe the approach and demonstrate it by the example of the problem of microelectronic design projects evaluation. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Musical harmonization with words: Realizability, potential issues and challenges

    Publication Year: 2009 , Page(s): 1 - 4
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (815 KB) |  | HTML iconHTML  

    Even though the number of publications regarding the application of the fuzzy logic to music is low, due to the nature of the music theory, it is appropriate to apply fuzzy based reasoning for the problems in music. These problems can be musical harmonization, improvisation, and even composition. The applicability of fuzzy logic to such problems is due to its power for the representation of human perception. Going to extremes, in this proceeding, we propose the idea of capturing the desire of the arranger/composer (which is stated in the natural language) and mathematical representation of these stylistic and personal preferences in an automatic manner via the concept of ¿computing with words¿. Although the idea seems to be realizable for simple scenarios, certainly there exist numerous potential issues and challenges especially for the complex ones. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Principal component based classification for text-independent speaker identification

    Publication Year: 2009 , Page(s): 1 - 4
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (419 KB) |  | HTML iconHTML  

    Classification based on Principal Component analysis has recently appeared in the literature in application to text-independent speaker identification. However, results have been reported for only clean speech data. In this paper, we evaluate the performance of principal component classifier for text-independent speaker identification on telephone speech. We then improve its identification performance using a Vector Quantization classifier in combination, through fusion of classifier scores. An identification rate of 78.27% has been obtained on the NTIMIT database, which is well above the best identification rate ever reported in the literature obtained by using only one type of feature set. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A new parallel programming language fortress: Features and applications

    Publication Year: 2009 , Page(s): 1 - 4
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1263 KB) |  | HTML iconHTML  

    An overview of a newly designed parallel programming language is presented in this paper. Fortress, an outcome of the high productivity computing systems program at DARPA, is examined through its features like mathematical notation, implicitly parallel nature and ease of library coding, its potential to grow easily and its interchangeable program parts. Code writing is discussed through examples. The tests for its scalability are done with several array types and also a comparison of an application written both in C and Fortress is presented. The usability and the future of the language in HPC environments are also discussed. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Economic order quantity model with backorders using trapezoidal fuzzy numbers

    Publication Year: 2009 , Page(s): 1 - 4
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (439 KB) |  | HTML iconHTML  

    Inventory is a common name stands for the items which are in stock points or work-in-process serving to decouple successive operations. Effective inventory management is essential for manufacturing organizations for a number of reasons. A basic economic order quantity model allowing backorders is considered in this paper. All cost parameters and the demand is assumed to be fuzzy numbers with trapezoidal membership function. By the graded mean integration representation method of defuzzification, total cost formula will be derived in the fuzzy sense in order to obtain the optimal order quantity. A numerical example is provided to illustrate the results of the proposed model. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • EOG controlled mobile robot using Radial Basis Function Networks

    Publication Year: 2009 , Page(s): 1 - 4
    Cited by:  Papers (4)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (393 KB) |  | HTML iconHTML  

    Controlling a mobile robot using human biopotential signals has been a common problem in the field of assistive robotics. Not only it is enough to analyze the biosignal characteristics and interpret motion commands from the raw signal, but also an efficient learning algorithm may help to overcome varying characteristics of the biosignal for the sake of robust control of the mobile robot. In this work, an efficient learning algorithm utilizing Radial Basis Function Networks have been studied and applied to EOG signals in order to control a mobile robot. Obtained results show that RBF network is successful in learning the biosignal characteristics and producing sufficient control signals to control a mobile robot. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Agent-based centralized fuzzy Kalman filtering for uncertain stochastic estimation

    Publication Year: 2009 , Page(s): 1 - 4
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (438 KB) |  | HTML iconHTML  

    In this paper, we investigate the problem of agent-based centralized fuzzy Kalman filters observing an uncertain physical process with parametric uncertainties. An agent-based sensor network is a distributed system which consists of sensors with limited computational capabilities. In our agent-based sensor network we consider sensor agents and a moderator agent. Any of these sensor agents have limited computational capabilities and also may be affected by different noises. Agents derive the information in the form of fuzzy states from their fuzzy Kalman filters, the estimated fuzzy states would be transmitted to the moderator agent for aggregation and result sharing by any sensor agent. The moderator agent fuses the fuzzy estimations to generate the global state estimations which is highly reliable. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Optimal query answering in fuzzy knowledge bases

    Publication Year: 2009 , Page(s): 1 - 4
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (202 KB) |  | HTML iconHTML  

    Under a fuzzy knowledge base (f.k.b.) we mean a finite set of formulas of the Zadeh's propositional fuzzy logic (based on a triangular norm min{x, y}). A sentence is a formula with lower and upper bounds - truth values from [0,1]. A state ¿ of f.k.b. S is the sentences set which is obtained by an assignment of lower and upper truth value bounds to each formula of S. A query ¿ to f.k.b state ¿ is a formula with atoms occurring in S. An answer to the query ¿ is a sentence a which logically follows from ¿. The optimal query answer is the answer with the closest bounds. We consider the problem of finding optimal query answers to f.t.b. states. The problem can be by analytical tableaux method. The method results in an algorithm with the exponential worst-case estimate (relatively to the size of ¿¿{¿}). But computation of optimal answers can be sped up considerably when S and ¿ are fixed and ¿ is an arbitrary state of f.k.b. S. Really, consider the general state ¿0 i.e. a state with bounds which are not specific values but parameters (indeterminate values). Then it is possible to find the general optimal answer a to query ¿ to the general state ¿0 This answer has the bounds that are expressions composed from the parameters, and any specific general answer can be obtained by assignments of specific values to the parameters. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Turkish spelling error detection and correction by using word n-grams

    Publication Year: 2009 , Page(s): 1 - 4
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (448 KB) |  | HTML iconHTML  

    N-grams can be used for spelling check and correction processes. The first step to use n-grams is to find the language specific n-grams by using a corpus. But a corpus cannot be big enough to contain all the possible word n-grams. Back-off smoothing technique is one of the techniques to estimate the frequency of the unknown n-grams in a corpus. By using Back-off technique and the Minimum Edit Distance (MED) algorithm, a program was developed to check spelling errors and suggest corrections in a sentence typed in Turkish. The results were compared with the results of Microsoft Word 2003 proofing tools, and found to be much better. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A new method for attribute extraction with application on text classification

    Publication Year: 2009 , Page(s): 1 - 4
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (473 KB) |  | HTML iconHTML  

    We introduce a new method for dimensionality reduction by attribute extraction and evaluate its impact on text classification. The textual contents in body sections of the news in Reuters-21758 are the selected attributes for classification. Using the offered method, high dimension of attributes- words extracted from the news bodies- are projected onto a new hyper plane having dimensions equal to the number of classes. Results show that processing times of classification algorithms dramatically decrease with the attribute extraction method we offer. This is achieved by the fall of the number of attributes given to classifiers. Accuracies of the classification algorithms also increase compared to tests run without using the proposed method. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Symmetry analysis for 2D images by using DCT coefficients

    Publication Year: 2009 , Page(s): 1 - 4
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (525 KB) |  | HTML iconHTML  

    In this study, we proposed a new method to align symmetric signals by using symmetry property of discrete cosine transform (DCT), which is widely used in signal compression and pattern recognition. For the symmetric signals, the energy is concentrated in the even indexed DCT coefficients. Using this property, we defined a symmetry measure. In this measure, ratio of the energy in even indexed coefficients to total energy gives the symmetry value for the signal. When the symmetry values of the rotated image at different angles become maximum, it means that the image is aligned according to its symmetry axis. We use 2D face data in experimental studies, because of its well-known symmetric property. This symmetry measure can also be adapted to any dimensional signals. The proposed method is tested on texture and shape data of the 3D face which is taken from FRGC database. Using the proposed method which exploits face symmetry, it is shown that the alignment resolution better than 1° can be achieved. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fitness landscape analysis of differential evolution algorithms

    Publication Year: 2009 , Page(s): 1 - 4
    Cited by:  Papers (2)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (358 KB) |  | HTML iconHTML  

    Fitness landscape analysis in evolutionary algorithms is commonly done on problems represented as bit strings with Hamming distance based random walks on the landscape. In this study, we aim to do a preliminary fitness landscape analysis of the differential evolution algorithm, which works on continuous search spaces. To the authors' best knowledge, no such fitness landscape analysis has been conducted in literature on continuous problems where search is performed through differential evolution. To achieve this aim, we first propose a suitable neighborhood definition through which a vector-based random walk on the landscape is possible. Then we use this neighborhood definition to conduct a fitness distance correlation and a correlation length analysis on a series of benchmark functions. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Experimental investigation of data transmission in wireless ad hoc networks

    Publication Year: 2009 , Page(s): 1 - 4
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (448 KB) |  | HTML iconHTML  

    The paper describes implementation of an application-layer program for experimental investigation of data transmission in wireless ad hoc networks in a real-word environment. The designed multithreaded program is executed on the laptops running the Vista operating system. The program performs routing and data transmission in different ad hoc network configurations. Some important performance metrics in wireless ad hoc networks are studied with this program. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A new adaptive fuzzy controller for DC motor position control

    Publication Year: 2009 , Page(s): 1 - 4
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (439 KB) |  | HTML iconHTML  

    Armature voltage controlling or field controlling is two usual methods for position control of DC motors. As reviewing of researchers work show in many controllers optimizing the controller parameters is a part of designing process. In this work, position of a DC motor is controlled by using scaling factor and switching between two fuzzy controllers automatically. In simulation, variations of load and armature voltage are applied and the result shows the controller overcome chattering successfully. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Optimal feature selection for 3D facial expression recognition with geometrically localized facial features

    Publication Year: 2009 , Page(s): 1 - 4
    Cited by:  Papers (2)  |  Patents (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (553 KB) |  | HTML iconHTML  

    The design of a recognition system requires careful attention to pattern representation and classifier design. Some statistical approaches choose these features, in a higher dimensional initial space, which allow sample vectors belonging to different categories to occupy compact and disjoint cluster regions in a lower dimensional subspace. The effectiveness of the subspace is determined by how well samples from different classes can be separated. This paper describes a feature selection process for a pose invariant 3D facial expression recognition method providing a lower dimensional subspace representation, which is optimized to improve the classification accuracy, retrieved from geometrical localization of facial feature points to classify universal facial expressions. Probabilistic neural network architecture is employed as a classifier to recognize the facial expressions from the feature vectors obtained from 3D facial feature locations. Facial expressions such as neutral, anger, disgust, fear, happiness, sadness, and surprise are successfully recognized with an average recognition rate of 93.72%. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.