Search

Snake-in-the-box is a graph theory problem of finding the longest path in a hypercube with some constraints. As the dimension of the hypercube increases, the number of paths grow, increasing the search space exponentially. In the past, genetic algorithms have been known to work well with problems having such large search spaces. In this paper, a genetic algorithm using a modification of the edge recombination operator is presented to find the longest snake in dimensions one to nine with lesser edge failures.

This paper introduces a novel approach to perform Diffuse Optical Tomography overcoming the limitation and high costs of Time Domain conventional methods, using a multichannel double stage coding technique. A preliminary numerical validation of the model is presented.

The aim of the paper is building an application that could ease the everyday life of a person affected by Alzheimer's disease. This application provides various functionalities such as tracking movements of the patient through GPS, providing medicine and food timing notifications, daily routine tracker and quiz to increase cognitive functioning of the patient. The main objective of this project is to make people suffering from Alzheimer's as independent as possible. So this project is not only about creating an application but also address the problem of increased dependency of Alzheimer patients on caregivers.

The fundamental idea behind Software Defined Networking (SDN) architectures is to decouple the control plane and data plane of the network to simplify the complexity of the network and centralize the control plane. A major feature of an SDN is the ability to have the controllers distributed over the network to solve the problem of a single point of failure at the controller, and to increase scalability and flexibility. Despite the flexibility and scalability of using multiple controllers, which facilitates the use of a larger number of network switches, yet the intercommunication cost between distributed controllers is still a key challenge in SDN. This paper presents an effective approach for the controller placement problem in software defined networks that aim to minimize the communication delay time between the controllers and the delay time between controller and switch. The major concept employed in this work is the Connected Dominating Set (CDS). Furthermore, an algorithm using the Minimum Connected Dominating Set (MCDS) is presented in which the delay time between the distributed SDN controllers is minimized. Simulation results are presented to show the efficiency of our approach compared to previous methods.

Cloud storage is a cloud-computing model in which data is stored on remote servers accessed from the internet. It has significantly changed the way users and administrators manage and access their data. Using remote storages to store data has many advantages in terms of availability and operational costs, but the security of such data is still one of the major concerns for the users. Pollution attack, where an adversary modifies some of the stored data is one of the many potent risks that affect the cloud data. In this paper, we show how disastrous pollution attack can be in coding based block level cloud storages, and how our algorithm using LRC, a version of Raptor codes, can identify an attack even before decoding all of the received packets.

3D garment model plays an important role in the fashion industry. However, not much work focus on extracting high-quality clothes from scanned data. The texture is limited by photography methods in 3D scanning. This paper proposes a novel framework of reconstructing high-quality 3D garment models with synthesized texture. Firstly, a pipeline of 3D garment processing is proposed to obtain a better 3D model based on KinectFusion. Then, DeepTextures is used to synthesize a new texture. To our best knowledge, this is the first paper combining 3D garment reconstruction and texture synthesis. Experimental results show that our method can conveniently obtain 3D garment models and realistic textures.

The paper is concerned with the coordination of interaction between various types of consumers and a power company to manage electricity consumption by using the adverse selection model based on contract theory. There are several power companies competing in the electricity market according to the Bertrand model. A method is proposed for load curve optimization by different types of consumers and a power company. Moreover, the types of consumers are identified and distinguished. The utility functions constructed for them describe the real situation rather well, and allow the implementation of a system of incentives for load curve optimization (load shifting from a peak time of the day by bounded rational consumer). The prices in rates providing a separating equilibrium are determined.

In this paper, we propose a mobile application that provides users with physical activity recommendations on a daily basis. The increasing percentage of physical inactivity among individuals [1] is the main motivation behind this system. Thus, this system represents one possible solution to counter this issue by recommending the minimum threshold of daily physical activity required for individuals to maintain an active lifestyle. The system performs this task by utilizing a biofeedback sensor (accelerometer) that tracks user's movements. Users’ calorie consumption is calculated from acceleration data and is then used as a foundation for system recommendations. Also, context data is utilized to provide context- aware recommendations, which will make the suggested activities from our system easier to follow for individuals. The proposed system gives recommendations in the forms of textual, audio and haptic recommendations. A questionnaire has been conducted in order to estimate the importance of the proposed system through estimating participants’ daily activity levels. Additionally, it aims to determine context data that can support the proposed system most fully. Results upheld the proposed system with 46% of participants who exercise for less than 15 minutes daily. Our findings also show that weather conditions and work commitments are the factors, which have the strongest negative impact on an individual's physical activity.

It is well known that data loss caused by data stolen Trojans is huge as it could upload privacy or secret data to hackers who controls it remotely. Most of current security tools monitor Trojans by scanning the signature code that is distinguished from normal software. However, this method can only recognize known Trojan except up-to-date malicious software that has unknown signature code. Some other security tools requiring preinstalled on hosts detects Trojans by program behaviors. This paper proposes a novel medel to detect data stolen Trojans based on their network behaviors. It consists of three detectors: 1) keep-alive detector detects keep-alive packets or connections; 2) master-slave-connection detector tries to find master and slave connections and 3) mistake detector analyses the rate of download vs. upload and connection time for different protocol. The experiments show that this method is efficient in recognizing data stolenTrojans. This protyped system proves the possibility of detection Trojans from network.

The highlight of this paper is to demonstrate the concept of adaptive tests which are an efficient way to produce the desired result as compared to the traditional static tests. The algorithm learns to adapt to the user's knowledge, and thus, calculates the true ability of the user. Our test system also provides the user with his performance review in specific categories, thereby giving him an understanding of his current knowledge for the same. Thus, our model can improve the user‟s efficacy over a prolonged period of time.

This paper presents the design and implementation of a lightweight node authentication protocol which comprises of node registration, node authentication and key establishment phases. The solution leverages the low computational overheads associated with cryptographically secure one-way hash chains and Elliptical Curve Crytography (ECC) without using any digital signature algorithm or any public key cryptography. The usage of hidden generator point derived from hash-chains provides defense against a man-in-the-middle attack which is prominent in ECDH (Elliptical Curve Diffie-Hellman) due to lack of entity authentication. The authentication protocol has been simulated on Tossim and its performance bench marking has also been carried out.

With the rapid development of e-commerce, online reviews have become one of the factors that affect the consumer purchasing decision. This paper proposes to use integrated Receipt-Accept-Sample(RAS) model and membership level to analyze how online reviews affect consumers’ purchasing attitude. The result shows that consumers with different membership levels often have different attitudes towards online reviews. Consumers with higher membership levels normally have a greater probability to receive online review information, but a lower probability to accept it.

In discovering the nature of the Primary Language of the human brain introduced by J. von Neumann, two components have been investigated. The first component is Linguistic Geometry (LG), the algorithm of optimizing war-fighting strategies. LG's universal applicability in various domains and its power in generating human-like strategies suggested that LG should be a component of the Primary Language. The second component is the algorithm of inventing new algorithms. This paper researches the role of mosaic reasoning, a component of the Algorithm of Discovery, in obtaining the key results in LG such as the algorithms for generating trajectories and zones.

This paper giving the associated credit risk model with the imperfect immunization scenario in the enterprises cluster based on assets-related under the framework of the small world network, we apply mean field theory and principle of epidemiology, to reveal the relation among associated credit risk contagion probability with imperfect immunization and asset-related in enterprises community network. The study has found that the threshold value of the associated credit risk contagion is reduced under the condition of the immune failure and immune invalidity, thereby, increasing the likelihood of related credit risk contagion; and the assets correlation between enterprises will influence the associated credit risk contagion and immune effect.

Now this era of Agriculture is treated as ICT mediated Market Led Agriculture Extension era. So in this Information revolution age we can’t able to survive without Information. As per R.T.I. act 2005 every person deserves his/her right for Information. So by giving respect to Information we should create a Revolution on Information by the enchanting mantra “Soochana se Samadhan”. India is second largest producer of commodities such as fruits and vegetables. One of the key issues, which require research, is the method by which we can reduce the post harvest loss, which is quite substantial at present. This would need design of cost effective, efficient, environment friendly storage system. Also, there is need for value addition to agricultural produce to maximize the agriculture return. This paper attempts to highlight the importance of ICT in improving marketing activities of retail business in agricultural areas in Indian economy. This paper also discusses vast potential of implementing the same in Indian agricultural business activities with some success stories, models for justification of the importance of ICT in Agriculture Retail Marketing.

Knowledge of patients’ location information (postal/zip codes) is critical in public health research. However, the inclusion of location information makes it easier to determine the identity of the individuals in the data sets. An efficient way to anonymize location information is through aggregation. In order to aggregate the locations efficiently, the data holder needs to know the locations’ adjacency information. A location adjacency matrix is big, and requires constant updates, thus it cannot be stored at the data holder's end. A possible solution would be to have the adjacency matrix stored on a cloud server, the data holder can then query the required adjacency records. However, queries reveal information on patients’ locations, thus, we need to privately query the cloud server's database. Existing private information retrieval protocols are inefficient for our context, therefore, in this paper, we present an efficient protocol to privately query the server's database for adjacency information and thus preserving patients’ privacy.

The deployment of wireless body sensor (WBS) in e-health systems represents an attractive solution due to the low-cost and the wearability of these devices. WBS(s) can be used for automatic detection of epileptic seizures which can considerably improve the patients’ quality of life. However, the main challenge that faces the sensor-based solution is the energy efficiency. In this context, this paper proposes a new energy efficient scheme for onset epilepsy seizure detection for non-invasive EEG using wireless body sensor networks. This new scheme is based on local features’ extraction using Haar wavelet transform of the segmented EEG signal. If the epilepsy seizure is detected, a notification is then transmitted to a remote server. The paper discusses the design of this scheme. It also presents the performances’ analysis for epilepsy detection and its energy efficiency when implemented in sensors. The results show that for 10 sec segments, the proposed scheme has sensitivity of 97% and it saves 46% sensor energy compared to full streaming of the raw EEG data to remote server.

This paper presents a software environment for a human-aware ambient agent providing support for a human performing an attention-demanding task. The agent obtains human attention-awareness by use of a dynamical model of human attention, gaze sensoring by an eye-tracker, and information about features of the objects in the environment. It has been implemented in a component-based, event-driven manner within the Adobe ® Flex ® environment, thereby also integrating the Tobii ® eye-tracker. It has been applied in a setup for a task where the human has to identify enemies and allies, and eliminate the enemies.

Emotion regulation is a vital psychological process that allows people to manage their own emotional states. Recent psychological research has highlighted the importance of flexibility in emotion regulation, such that people can alternative between different emotion regulation strategies. A strategy is chosen depending upon the demands of the situation. This means that healthy emotion regulation is context-sensitive. This paper presents a computational model which models this form of flexible adaptation in emotion regulation in a simplified scenario in which the person has to switch between expressive suppression and attention modulation in managing anger in different work situations. Simulation results are reported that illustrate the capacity of the model to display adaptivity in emotion regulation across different contexts.

Semantic search is qualified – by web-related enterprises as well as, academic research – as a key technology, ensuring important improvements in terms of shared data understanding, while it leads to refined and targeted interpretations. Accordingly, ontologies are the focal asset for a well-functioning semantic search approach, since their ability to share, represent and reuse explicit and semantic domain specification. Nowadays, a multitude of ontologies containing up to hundreds of thousands of concepts are proposed. Thus, our challenge as researchers exceeds conceptualizing or creating ontologies to being able to choose the fitting and suitable one, taking into account specific criteria. This paper comes within the same context as it presents a novel approach for weighting OWL ontologies, in order to choose the most appropriate one from a set of proposed ontologies. Our approach takes into account not only the taxonomic structure, but also the semantic aspect of the ontology. Furthermore, semantic relationships and specific concepts are the favored since they reflect the semantic richness of the ontology.

Heterogeneous network (HetNet) is considered as a prime way to solve the limits of system capacity and broadband service coverage in traditional network. However, the deployments of small cells with varied sizes make the network topology more complicated. Self-organizing network (SON) technology, aiming to reduce the operational costs, is a significant technology in HetNet. One of the common use cases is to improve handover performance. In this paper, a handover optimization algorithm based on enhanced mobility state estimation (EMSE) is proposed. Considering both user equipment (UE) speed and handover types, the optimization algorithm based on EMSE combines selective Time-to-Trigger (TTT) and dynamic handover margin (HM)-adjusting in SON. Furthermore, the algorithm performance is compared with two different reference cases. Simulation results show that total handover failure has an obvious decline with our self-optimizing algorithm. Therefore, handover performance gets improved and UEs have better mobility robustness in HetNet through our algorithm.

Agent-based simulation lends itself to study emergent behaviour when agents respond autonomously, based on their unique and individual attributes, to external interventions. In transport, the vehicle-specific attribute of interest is its fuel efficiency and emissions. While research showed that vehicle-specific emissions can be accounted for in terms of pollutant quantities, this paper aims to demonstrate that the chosen model is sensitive to a vehicle’s designated emissions concept. In urban freight or city logistics, such behavioural sensitivity is necessary to evaluate interventions like low emission zones where access is provided to goods vehicles based on their emission category.

Comet assay is a well known method for evaluating DNA damage in individual cells. Silver stained comet assay images are employed in this study as they are widely used in clinical applications. When compared with fluorescent images silver stained images require additional pre-processing stages to effectively identify the region of interest. An improved method for effective automatic identification of true comets from the noisy silver stained images is proposed in this paper. The performance of the proposed method is compared with the recent related methods and shows better results in terms of %TP, %FN, %FP, Positive Predictive Value and sensitivity.

Inference problems in dynamically data-driven application systems use physical measurements along with a physical model to estimate the parameters or state of a physical system. Developing parallel algorithms to solve inference problems can improve the process of estimating and predicting the physical state of a system. Solution to inference problems using the variational approach require multiple evaluations of the associated cost function and gradient, where the gradient is defined as the increase/decrease inflection point of the variable between two points. In this paper we present a scalable algorithm based on augmented Lagrangian approach to solve the variational inference problem. The augmented Lagrangian framework facilitates parallel cost function and gradient computations. We show that the methodology is highly scalable with increasing problem size by applying it for the Lorenz-96 model.

Many crimes are committed where the only record of the event is in the memory of a witness or victim. Recovering a recognisable image of the offender’s face is then crucial for solving the crime. Traditionally, eyewitnesses describe the offender’s face and select individual facial features–eyes, hair, nose, etc.–to build a ‘composite’. This image is then published in the media so that someone can recognise it and phone the police with a name. Unfortunately, when tested using life-like procedures, this method rarely produces recognisable images. The current paper describes these systems for extracting such biometric information from witnesses. It also describes how useful they are and explores three such approaches for improving their effectiveness. Included are a new method to interview witnesses (a holistic-cognitive interview), a new method to present images to the public (animated composite) and a new system to construct the face (EvoFIT).

This research is motivated by the need for face recognition in uncontrolled environments. In other words, we are interested in face recognition arrangements whereby the users do not need to interact with the recognition technology. The contribution of this paper is to perform a range of recognition experiments on face image data as people casually enter a building, without any instructions about expression. Specifically, we capture four images per session in rapid succession (all within 20 ms). The four images are synchronised to different light sources to enable photometric stereo processing to estimate albedo images, surface normals and depth maps. Additional capture sessions then take place over periods of many weeks. Our recognition experiments are on each of the three modalities as well as a fusion technique for the albedo and depth. Using a variety of photometric stereo methods, surface integration methods (to recover depth) and recognition algorithms such as principal component analysis and nonnegative matrix factorisation, we acquire a maximum recognition rate of 86% for 96 subjects.

A major consideration in state-of-the-art face recognition systems is the amount of data that is required to represent a face. Even a small (64×64) photograph of a face has 212 dimensions in which a face may sit. When large (>1 MB) photographs of faces are used, this represents a very large (and practically intractable) space and ways of reducing dimensionality without losing discriminatory information are needed for storing data for recognition. The eigenface technique, which is based upon Principal Components Analysis (PCA), is a well established dimension reduction method in face recognition research but does not have any biological basis. Humans excel at familiar face recognition and this paper attempts to show that modelling a biologically plausible process is a valid alternative approach to using eigenfaces for dimension reduction. Using a biologically inspired method to extract the certain facial discriminatory information which mirrors some of the idiosyncrasies of the human visual system, we show that recognition rates remain high despite 90% of the raw data being discarded.

This paper relates to a face recognition and verification technique based on ridgelet transforms. Our proposed method first uses the ridgelet transform of the face image for feature extraction. This involves first applying a sequen- tial combination of radon and wavelet transforms to both the training and test images. The result is then decomposed into a set of feature vectors. The Euclidean distance between training and test feature vectors is finally used for the actual recognition. Before applying any transform on the training image, we normalize the image using a segmentation process based on the YCbCr colour model. This essentially detects the largest region of skin in the image. Experimental results using Yale, AT&T, faces94, faces96 and Indian databases show the superiority of the proposed method with some of the existing popular algorithms.

Face recognition is one of the challenging applications of image processing. Robust face recognition algorithms should posses the ability to recognize identity despite many variations in pose, lighting and appearance. Principal Component Analysis (PCA) has been widely adopted as a potential face recognition algorithm. However, it has limitations like poor discriminatory power and large computational load. In view of these limitations with PCA, this paper proposes a face recognition method with PCA based on Gabor features. On applying the statistical models like Independent Component Analysis (ICA) and Linear Discriminant Analysis (LDA) on the output of reduced features from PCA, the more discriminating features were obtained. Two normalization methods, namely Unit Length normalization (UL) and zero Mean and unit Variance (MV) methods were employed for the normalization of extracted features in order to get a better classification results. The proposed Gabor feature based method has been successfully tested on ORL face data base with 400 frontal images corresponding to 40 different subjects which are acquired under variable illumination and facial expressions. It is observed from the results of PCA with Gabor filters that the ICA method gives a top recognition rate of about 95% when compared to LDA method with MV normalization method.

Face recognition under varying illumination is one of the challenging problems in real-time applications. Numerous methods have been developed by the research community to handle the problem. Existing surveys of methods are either too old or do not cover performance analysis of illumination invariant methods. This paper is more extensive than previous surveys and covers recently developed methods. The paper focuses on passive methods which solve the illumination problem by investigating the visible light images in which the face appearance has been altered by varying illumination. The methods are classified into four broad categories, namely (1) subspace-based statistical methods (2) illumination invariant representation methods, (3) model based methods, (4) other illumination handling methods. The other illumination handling category includes the methods which do not fall under first three categories. Performance analysis and discussion of methods and an evaluation of results is presented to determine the suitability and applicability of the method(s) for specific applications. It is observed from the survey of methods that illumination invariant representation based methods are better in terms of the number of training images required, the simplicity, computational complexity and robustness.

This paper presents a fast and reliable approach for the localization of an iris image using neural networks by comparing some of the existing localization methods. Normally any biometric recognition system is used for identification of an individual and verified with the available database to check whether the person is authorized or not. Nowadays, iris recognition systems are considered the best authentication method compared to other biometric systems due to the unique characteristic feature of the human iris. In iris recognition systems, the important and difficult step is to locate or segment the iris from the input eye image which is responsible for success rate of the iris recognition. Hence this paper suggests a efficient approach to locate the iris using neural networks in order to improve the efficiency of recognition systems. Further, the paper compares the existing iris localization methods such as Daugman algorithm, Hough transform and Canny edge detector algorithm. The best localization algorithm is chosen for training a network and simulating for efficient and fast segmentation of irises with good success rate.

Biometric based identification is an emerging technology that can solve many security problems. Although systems based on fingerprint and eye features have so far achieved the best matching performance, human hand also contains a wide variety of features, e.g. shape, texture and principal palm lines etc. This feature of the human hand is quite stable and hand images can be extracted very easily. Palm-print is a new and emerging biometric feature for personal recognition. This paper describes an automated approach to palm-print recognition. In contrast to existing palm-print based biometric systems, a new system which is resistant to motion variance of the palm has been proposed. A Palm-print image is taken as an input and then a low pass Gaussian filter is applied to remove the noise from image. Further, since the centre of mass always remains constant, so this property has been taken into account to extract the ROI (Region of Interest) from palm-print image. SIFT (Scale Invariant Feature Transformation) features have been used further for extracting stable texture features from the ROI and are stored. These stable features extracted from the ROI are further used for comparison with stable texture features extracted from ROI of other palmprint images to provide biometric based identification and security.

In this paper, we present a comparative analysis of Multi-algorithmic and Multimodal approaches. We have used palmprint and face as biometric traits and other popular subspace algorithms (PCA, FLD, and ICA). Subsequently, the different combinations of algorithms are also evaluated in our experiment. The multi-algorithmic approach achieves incremental results where as multimodal approach yields far improved results. Hence Complimentary information available through multimodal approach always performs better than multi-algorithmic approach which mainly builds on supplementary information.

This paper presents a secure and robust watermarking scheme to enhance the security of biometric template over insecure network. Diffusion and digital watermarking techniques are used to improve the security and secrecy of the templates. Diffusion phase is done by chaotic sequence and Hessenberg decomposition. This phase essentially change the pixel values of biometric template randomly. Finally, the watermark image, which is the face image of the owner of biometric template, is embedded in the diffused biometric template with the help of singular value decomposition. The feasibility of the proposed method and its robustness against different kind of attacks are verified by computer simulations.

In this paper we have proposed a technique for securing biometric traits using the modified Hill Cipher with an involutory key and a robust cryptosystem. The Modified Hill Cipher solves the drawbacks of conventional Hill ciphers by using iterations and interlacing. Interlacing of the binary bits of the plaintext image creates confusion while iterations create complexity to the cipher image. The Modified Hill cipher is secure to known plaintext. Binary conversion is only possible for integer elements so an integer involutory key matrix is used. As the key matrix is involutory, which eliminates necessity of matrix inverse while decryption. We have used a robust cryptosystem algorithm to send the cipher matrix safely. The cryptanalysis and histogram graphs show the variation of image property before and after encryption.

Multicast key distribution in mobile adhoc networks is challenging due to its inherent characteristics of infrastructure-less architecture with lack of central authority and limited resources such as bandwidth, time and power. In many multicast interactions, a new member can join, a current member can leave at any time and existing members must communicate securely using multicast key distribution within constrained energy for mobile adhoc networks. This paper proposes a new efficient cluster based multicast tree (CBMT) algorithm for secure multicast key distribution, in which the source node uses a Multicast version of the Destination Sequenced Distance Vector (MDSDV) routing protocol to consume less energy. It also reduces end to end delay in multicast transmission. Simulation results show the efficient CBMT using MDSDV have better system performance in terms of QOS metrics such as energy, latency, key delivery ratio and packet drop rate under varying network conditions. This proposed scheme achieves reliability, while exhibiting low packet loss rate with high key delivery ratio.

Mammography is a widely used diagnostic technique for early breast cancer detection in women. Clusters of Microcalcification are the sign of breast cancer and their detection will decrease the probability of mortality rate and improves its prognosis. The detection of microcalcification clusters is a difficult task for radiologists because of variations of size and orientation and are highly correlated with background tissue. In this paper, we present a Computer Aided Detection (CAD) method, which is used to detect nodules (microcalcification) in mammograms. We have designed a multi-scale filter bank based on the concept of second-order partial derivatives (Hessian matrix). Regions Of Interest (ROI) are identified by a multiresolution based histogram technique. This ROI of mammogram is decomposed into sub-bands, the low-frequency subband is suppressed and then the high-frequency subbands which contain only nodule-like structures are reconstructed. This structure is determined by the eigenvalues of the Hessian matrix. The detection performance of the proposed method is evaluated by comparing our results with two traditional wavelet based methods. Experimental results show that the microcalcifications can be efficiently detected by proposed method and it has high true positive ratio in comparison to other methods.

Epilepsy is a disorder of the central nervous system, specifically the brain. It is a neurological malfunction affecting about 1% of the population and is the third most common neurological disorder following rheumatic heart disease and Alzheimer’s disease, but it imposes higher costs on society. Magnetic Resonance Imaging (MRI) is one of the most common diagnostic tests used for patients for epilepsy prediction. Shortage of radiologists and the large volume of MRI scan images that need to be analyzed may lead to labor intensive, expensive and inaccurate prediction. Hence there is a need to generate an efficient prediction model for making a correct diagnosis of epilepsy and accurate prediction of its type. This paper describes the modeling of epilepsy prediction using Support Vector Machines (SVM), a machine learning algorithm. The prediction model has been generated by training the support vector machine with descriptive features derived from MRI data of 350 patients and observed that the SVM based model with a Radial Basis Function (RBF) kernel produces 93.87% of prediction accuracy.

One of the important issues in the design of fuzzy classifiers is the formation of fuzzy if-then rules and the membership functions. This paper presents a hybrid Particle Swarm Optimization based approach for fuzzy classifier design which incorporates the concept of mutation from evolutionary computations. The proposed MutPSO develops the fuzzy classifier system by encoding and evolving both the membership functions and rule set as particles simultaneously. Non-uniform mutation is applied to the membership functions which are represented as real numbers. Uniform mutation is applied to the rule set which is represented as discrete numbers. In the classification problem under consideration, the objective is to maximize the correctly classified data and minimize the number of rules. This objective is formulated as a fitness function to guide the search procedure to select an appropriate fuzzy classification system so that the number of fuzzy rules and the number of incorrectly classified patterns are simultaneously minimized. The performance of the proposed MutPSO approach is demonstrated through development of fuzzy classifiers for iris data available in UCI machine learning repository. Simulation results show the suitability of the proposed approach for developing the fuzzy system.

Image registration is of great significance to medicine and remote sensing, so a lot of techniques have been developed within the context of one or the other discipline. This paper proposes an approach for medical image registration using Modified Gabor Wavelet Transform (MGWT) for Modified Adaptive Polar Transform (MAPT). This algorithm can be used to register images of the same or different modalities. This transform analyzes periodic signal components and presents the advantage of being independent of the window length. The performance of the Modified Gabor Wavelet Transform is compared with previous methods like Log Polar Transform and Adaptive Polar Transform. The results show that MGWT outperforms all evaluated model-independent methods with respect to identification accuracy. These results show that the basis of errors produced by the previous methods is the fixed working scale. The new method not only avoids this basis of errors but also makes a tool available for detailed study.

At low bitrate and with acceptable quality in Fractal Image Compression (FIC) of the coded image, the encoding time is very large for most existing algorithms. In this paper, a fast fractal encoding system is proposed using particle swarm optimization (PSO) to reduce the encoding time. Here, an optimization technique is used for the MSE based on the stopping criterion between range block and domain block. This PSO technique can speedup the fractal encoder and preserve the image quality for medical imaging.

Images and text form an integral part of website designing. Images have an engrossing appeal and that’s why they attract more and more visitors. But, due to expensive bandwidth and time-consuming downloads; it has become essential to compress images. There are various methods and techniques available to compress images. In this paper, an effective technique is introduced called Wavelet-Modified Single Layer Linear Forward Only Counter Propagation Network (MSLLFOCPN) technique to solve image compression. This technique inherits the properties of localizing the global spatial and frequency correlation from wavelets. Function approximation and prediction are obtained from neural networks. Consequently counter propagation network was considered for its superior performance and the research helps to propose a new neural network architecture named single layer linear counter propagation network (SLLC). Several benchmark images are used to test the proposed technique combined of wavelet and SLLC network. The experiment results when compared with existing and traditional neural networks shows that picture quality, compression ratio and approximation or prediction are highly enhanced.

This paper presents a functional approximation of the M/D/1/N built on a Taylor series approximation. Numerical examples are carried out to illustrate the performance of our approach.

Recommender systems are widely used in many areas, especially in e-commerce. Recently, they are also applied in e-learning tasks such as recommending resources (e.g. papers, books,..) to the learners (students). In this work, we propose a novel approach which uses recommender system techniques for educational data mining, especially for predicting student performance. To validate this approach, we compare recommender system techniques with traditional regression methods such as logistic/linear regression by using educational data for intelligent tutoring systems. Experimental results show that the proposed approach can improve prediction results.

In this paper we propose a method for personalized recommendation of assignments (tasks or exercises) in an adaptive educational system. Our main goal is to help students to achieve better performance in tests. To achieve this we enhance existing adaptive navigation approaches by considering the limited time for learning. Our strategy is to cover all the required topics at least to some extent rather than learn few topics perfectly. The proposed method uses utility-based recommending and conceptbased knowledge modeling. We evaluate our approach in the domain of learning programming.

Collaborative filtering (CF) techniques have proved to be effective in their application to e-commerce and other application domains. However, their applicability to the recommendation of learning resources deserve separate attention as seeking learning resources can be hypothesized to be substantially different from selecting information resources or products for purchase. To date there are only a few scattered studies reporting on the application of well known user-based CF algorithms to learning object repositories. This paper reports an empirical study carried out by using MERLOT data and existing user-based CF algorithms. The aim of this preliminary study was that of finding evidence on accuracy measures of existing CF algorithms, and the relation of the items recommended with other elements of the repository. The results can be used as a starting point for future studies that account for the specific context of learning object repositories and the different aspects of preference in learning resource selection.

An educational digital library may store a wealth of diverse educational resources targeting different audiences from young schoolchildren to graduate students to school and college teachers. With the growth of the volume and the diversity of the library, it becomes increasingly difficult for the users to find resources, which are relevant to their age, educational needs, and personal interests. Social navigation techniques could provide valuable help in this context guiding users to the most useful information. Social navigation works by processing traces of past user behavior and using the assembled “collective wisdom” to guide future users. The paper reports our work on the design of social navigation infrastructure for Ensemble, a major educational digital library. We present the organization of both sides of the social navigation process: how social wisdom is collected and how it can be used to guide portal users to valuable resources. We also report the results of our most recent evaluation of the social navigation infrastructure.

Iris recognition a new biometric technology has great advantages such as variability, stability and security. In this paper we propose a feature extraction method for iris recognition based on contourlet transform. Contourlet transform captures the intrinsic geometrical structures of iris image. It decomposes the iris image into a set of directional subbands with texture details captured in different orientations at various scales. Discriminant analysis is used to determine the optimal threshold for the selection of dominant directional energy components. Only dominant directional energy components are employed as elements of the input feature vector. These input feature vectors are compared with the template feature vectors. Experimental results show that the proposed method reduce processing time and increase the classification accuracy and outperforms the wavelet based method.

Macroscopic and microscopic modeling have become mainstream methodologies for crowd simulation in dynamic environments. The two models make a trade-off between efficiency and accuracy, but neither of them is able to achieve both goals at the same time. With the aim of achieving both efficiency and accuracy, a hybrid modelling method is proposed in this paper for crowd simulation. This paper illustrates how the two types of models co-exist in a single simulation and work collaboratively. A case study for this method is also conducted, the simulation result of which shows that the proposed method can not only benefit from the macroscopic model by improving the simulation efficiency, but also obtain a fine-grained simulation result by adopting the microscopic model.

Goal oriented dual weight error estimation has been used in context of computational fluid dynamics for several years. The adaptation of this method to geophysical models is the subject of this paper. A differentiation-enabled prototype of the NAG Fortran compiler is used to generate a discrete adjoint version of such a geophysical model and allows to compute the required goal sensitivities. Numerical results are presented for a shallow water configuration of the Icosahedral Non-hydrostatic General Circulation Model (ICON). A special treatment of the underlying linear solver is discussed yielding improved scalability of this approach and a significant reduction in memory consumption and runtime.