ERU - 2016

Permanent URI for this collectionhttp://192.248.9.226/handle/123/19525

Browse

Recent Submissions

Now showing 1 - 14 of 14
  • Thumbnail Image
    item: Conference-Abstract
    A simplified epilepsy classification technique utilizing svd
    (Engineering Research Unit, Faculty of Engiennring, University of Moratuwa, 2016-04) Prabhakar, SK; Rajaguru, H; Jayasekara, AGBP; Amarasinghe, YWR
    EEG signals represent both the brain function and also the status of the whole body, i.e. a simple action as blinking the eyes introduces oscillation in the EEG records. The EEG is a direct way to measure neural activities and it is important in the area of biomedical research to understand and develop new processing techniques. EEG signal pre-processing and postprocessing methods include EEG signal modeling, segmentation, filtering and de-noising, and EEG processing methods which consist of two tasks, namely, feature extraction/dimensionality reduction and classification. In this paper, the performance analysis of Independent Component Analysis (ICA) is considered as a dimensionality reduction technique followed by Singular Value Decomposition (SVD) as a Post Classifier for the Classification of Epilepsy Risk Levels from EEG Signals. The analysis is done in terms of bench mark parameters such as Performance Index (PI), Quality Values (QV), Sensitivity, Specificity and Time Delay.
  • Thumbnail Image
    item: Conference-Abstract
    Short-term traffic prediction with visitor location registry data
    (Engineering Research Unit, Faculty of Engiennring, University of Moratuwa, 2016-04) Dilhasha, F; Fernando, K; Godahewa, R; Ossen, S; Perara, AS; Walpola, M; Jayasekara, AGBP; Amarasinghe, YWR
    Increasing road traffic is a major issue in current world. In this paper, we propose a set of prediction models that can perform short term traffic prediction for a given road segment. These prediction models have been developed using Neural Networks (NN), Bayesian Networks, Hidden Markov Models, variations of Regression and ensemble approaches of these models. CCTV records are used for validation of the results based on which a maximum accuracy of 85% was achieved.
  • Thumbnail Image
    item: Conference-Abstract
    Modeling and forecasting opec reference basket crude oil prices using artificial neural networks
    (Engineering Research Unit, Faculty of Engiennring, University of Moratuwa, 2016-04) Mahanthege, SR; Chandrasekera, NV; Jayasundara, DDM; Jayasekara, AGBP; Amarasinghe, YWR
    With Organization of petroleum exporting countries (OPEC) being the dominant player in the crude oil industry, it is of great importance to accurately forecast OPEC reference basket (ORB) crude oil prices. This study focuses on forecasting ORB crude oil prices using two Artificial Neural Network (ANN) models: Feedforward Neural Network (FFNN) and Time-Delay Neural Network (TDNN). Training of the networks were done by changing several parameters of the networks. Levenberg-Marquardt training algorithm has been used in training ANN. Results indicates that the TDNN model outperforms the FFNN model in forecasting daily ORB crude oil prices.
  • Thumbnail Image
    item: Conference-Abstract
    Human action detection using space-time interest points
    (Engineering Research Unit, Faculty of Engiennring, University of Moratuwa, 2016-04) Sriashalya, S; Ramanan, A; Jayasekara, AGBP; Amarasinghe, YWR
    The bag-of-features (BoF) approach for human action classification uses spatio-temporal features to assign the visual words of a codebook. Space time interest points (STIP) feature detector captures the temporal extent of the features, allowing distinguishing between fast and slow movements. This study compares the relative performance of action classification on KTH videos using the combination of STIP feature detector with histogram of gradient orientations (HOG) and histograms of optical flow (HOF) descriptors. The extracted descriptors are clustered using K-means algorithm and the feature sets are classified with two classifiers: nearest neighbour (NN) and support vector machine (SVM). In addition, this study compares actionspecific and global codebook in the BoF framework. Furthermore, less discriminative visual words are removed from initially constructed codebook to yield a compact form using likelihood ratio measure. Testing results show that STIP with HOF performs better than HOG descriptors and simple linear SVM outperforms NN classifier. It can be noticed that action-specific codebooks when merged together perform better than globally constructed codebook in action classification on videos.
  • Thumbnail Image
    item: Conference-Abstract
    Identify whether the user is in a meeting or not by using naïve bayes classifier
    (Engineering Research Unit, Faculty of Engiennring, University of Moratuwa, 2016-04) Yogaraja, AA; Dharmasena, HSR; Gobinath, L; Vishnuvathsasarma, B; Walpola, MJ; Jayasekara, AGBP; Amarasinghe, YWR
    This paper is discussed about implementing a Naïve Bayes classifier to identify whether the user is in a meeting or not for the context aware systems. We implemented an application (AutoProfile App) to identify whether the user is in a meeting or not by using a Naïve Bayes classifier.
  • Thumbnail Image
    item: Conference-Abstract
    Integrating context-awareness with reminder tools
    (Engineering Research Unit, Faculty of Engiennring, University of Moratuwa, 2016-04) Ranasinghe, YS; Walpola, MJ; Jayasekara, AGBP; Amarasinghe, YWR
    People use reminders to recall tasks to-do. But sophisticated enough tools are unavailable. Context-Awareness in mobile computing is more engaging area in gathering information about the user’s current situation and would be ideal to integrate with the reminders to generate context-aware reminder where rich context is used to identify when a reminder should be presented to its recipient. Here we have proposed a model to integrate context-awareness with reminder’s decision making process and develop a Context-Aware reminder.
  • Thumbnail Image
    item: Conference-Abstract
    Automated optimization and suggestion of libraries and patterns for software architectures
    (Engineering Research Unit, Faculty of Engiennring, University of Moratuwa, 2016-04) Pirapuraj, P; Perara, I; Jayasekara, AGBP; Amarasinghe, YWR
    It is clear that code reuse is important task in software development and maintenance. The problem in code reuse is, after download all relevant code, we need to identify most relevant code among those. In this paper we use keyword search with n-gram NLP technique using GitHub Application Program Interface (API). Before search the source code, we retrieve all Repository name in GitHub belongs to particular programing language (JAVA, C++, etc.), as well as we retrieve all .java file name if we search java libraries using GitHub API. Then we compare our keyword with this list, if the keyword is extracted from Software architecture is connected word, then we will split using Apache Camel Splitter. If the particular keyword related to any project, we download the project. Otherwise using WordNet, we use synonym and repeat above process again. For further relevancy, we will use a speech recognition technique (Dynamic Time Warping (DTW)) and a NLP technique (Part of Speech Tagging (POS)). Because of this is a part of the whole research, in this paper we will consider only GitHub API use.
  • Thumbnail Image
    item: Conference-Abstract
    Fabrication and characterization of cu2znsns4 thin films by spin coating
    (Engineering Research Unit, Faculty of Engiennring, University of Moratuwa, 2016-04) Wijewantha, NLW; Wijesinghe, DPS; Attygalle, D; Jayasekara, AGBP; Amarasinghe, YWR
    This study reports the fabrication and characterization of Cu2ZnSnS4 (CZTS) thin films which is inexpensive and environmental friendly photovoltaic (PV) material using solution based method. The CZTS thin films were coated on soda lime glass substrates using spray pyrolysis and spin coating methods. The thickness variation, electrical properties and optical properties of the films were determined for different numbers of coatings. The optical band gap energy of the prepared films was within the range of 1.5eV to 2.07eV. A set of prepared CZTS films on glass substrates were subjected to an annealing process in a normal atmospheric environment to obtain better crystalline properties [2]
  • Thumbnail Image
    item: Conference-Abstract
    Abstract of finding ideal conditions for azolla pinnata
    (Engineering Research Unit, Faculty of Engiennring, University of Moratuwa, 2016-04) Priyadarshani, EAC; Dissanayaka, NP; Kodikara, S; Jayasekara, AGBP; Amarasinghe, YWR
    The economic and environmental costs of heavy use of chemical fertilizers in agriculture are a global concern. Sustainability considerations mandate that alternatives to Nitrogen fertilizers must be urgently sought. Biological nitrogen fixation (BNF), a microbiological process which converts atmospheric nitrogen in to a plant usable from, offers this alternative. Nitrogen fixing systems offer an economically attractive and ecological sound means of reducing external inputs and improving internal resources. Symbiotic systems such that of Anabaena – Azolla system can be used as a source of nitrogen in cropping systems.
  • Thumbnail Image
    item: Conference-Abstract
    Bed shrinkage in packed bed biomass combustion
    (Engineering Research Unit, Faculty of Engiennring, University of Moratuwa, 2016-04) Adassuriya, TL; Jayatillake, KDKY; Nupehewa, JM; Kaluhendiwela, YC; Narayana, M; Jayasekara, AGBP; Amarasinghe, YWR
    This paper focuses on evaluating shrinkage of a biomass packed bed. A laboratory scale packed bed is designed to conduct experiments in batch mode. Different types of biomass i.e. Ginisiriya, coconut shells and rubber are used for the packed bed to study the behavior of bed shrinkage. The model uses inlet air stream at room temperature as the gasifying medium and a fixed batch of biomass. Ignition of the biomass batch is initiated by an external heat source. Further operation is maintained by the own heat emitted by combustion reactions, until the whole batch is finished. As the time goes height of the packed bed is decreasing due the shrinkage of the bed. Empirical data obtained from this model can be used to evaluate the bed shrinkage and temperature variations in solid phase and gas phase with time.
  • Thumbnail Image
    item: Conference-Abstract
    Non-destructive evaluation of strength gain of ordinary portland cement mortar by ultrasonic pulse velocity method
    (Engineering Research Unit, Faculty of Engiennring, University of Moratuwa, 2016-04) De Silva, SD; Sivahar, V; Jayasekara, AGBP; Amarasinghe, YWR
    This paper presents results of the study of ultrasonic pulse velocity and compressive strength of ordinary Portland cement mortar. Objectives of this study are to determine correlations between ultrasonic pulse velocity and compressive strength of cement mortar and to predict strength of cement mortar at 28 days within 02 days using ultrasonic pulse velocity of cement mortar. Water/cement ratio, sand content, method of specimen preparation, curing conditions were kept constant throughout the experiment. Portable Ultrasonic Nondestructive Digital Indicating Tester with transducers of 54 kHz frequency was used to measure ultrasonic pulse velocity. In this research it was found that the relationship between ultrasonic pulse velocity and compressive strength was linear and Lime Saturation Factor (LSF) of cement has an influence on this relationship. Further it was seen that, with the increase of curing period ultrasonic pulse velocity of all samples increased. At the end it was possible to obtain an equation that can predict the cement strength at 28 days within 02 days and that equation was verified using another five (05) cement samples.
  • No Thumbnail Available
    item: Conference-Extended-Abstract
    Development of a badminton shuttlecock feeding machine
    (Engineering Research Unit, Faculty of Engiennring, University of Moratuwa, 2016-04) De Alwis, G; Dehikumbura, C; Konthawardana, M; Lalitharatne, TD; Dassanayake, VPC; Jayasekara, AGBP; Amarasinghe, YWR
  • Thumbnail Image
    item: Conference-Abstract
    Design and implementation of a robot hand for service purposes
    (Engineering Research Unit, Faculty of Engiennring, University of Moratuwa, 2016-04) Pathirana, MGK; Perara, KTK; Sirithunge, HPC; Srimal4, PHDAS; Jayasekara, AGBP; Jayasekara, AGBP; Amarasinghe, YWR
    The paper focuses on designing and implementation of an interactive robotic arm for pick and place and service applications. The robot is expected to have image processing and voice recognition capability. Therefore once an object is placed on the coordinate plane of the robot, it has the ability to detect the colour or shape of the body, calculate its coordinates and then lift the object. The manipulator moves the end effector to the final position using inverse kinematics calculation of each joint angle. The mechanical structure is simulated using forward kinematics and for this Denavit-Hartenberg parameter method is used for the proposed design and hence the mechanical structure is simulated in MATLAB and SolidWorks. Forward kinematics analysis is done through Robotics Toolbox and Matlab. Manipulator is implemented according these parameters chosen according to the design requirements and assessed the scope of the robot arm.
  • Thumbnail Image
    item: Conference-Abstract
    Use of dipanalyst 2.0 software to analyse road cut slope failures
    (Engineering Research Unit, Faculty of Engiennring, University of Moratuwa, 2016) Premasiri, HMR; Senarathna, CSS; Godaliyadda, DP; Disanayaka, KTD; Jayasekara, AGBP; Amarasinghe, YWR
    This study the causes for the road cut failures and suggest a methodology to design road orientation to minimize cut failures considering geology and structure of the rocks present along road cuts. So this study is more important to minimize and control road cut failures while designing road orientation in scientific manner. This study is help to create a forecasting model for different scenarios. In this study to validate the outcomes some software’s were use specially DIPANALYST 2.0. It is a is slope stability analysis software, which is designed to perform kinematic analysis for rock slopes and also calculating factor of safety values for plane and wedge failures based on the limit equilibrium theory. Using the software all possible failures in the study area were analyzed and it can be further justify using other geological properties of rocks such as mineralogical composition.