ITRU - 2014

Permanent URI for this collectionhttp://192.248.9.226/handle/123/14726

Browse

Recent Submissions

Now showing 1 - 10 of 10
  • item: Conference-Full-text
    A Study on SSD Aware Scan Operation Optimization in PostgreSQL Database
    Nakandala, S; Withana, S; Kumarasiri, D; Jayawardena, H
    Cost functions used by query optimizers in database systems are built around the properties of conventional hardware technologies like spin type hard disk drives. But with the rapid development in storage technology several new technologies have emerged to overcome the limitations in conventional storage media. One such storage technology is flash storage type which belongs to the broader category of SSDs (Solid State Devices). In this paper we discuss a study done to optimize scan operation performance in PostgreSQL database running on flash storage media by changing the random block transfer cost parameter to a suitable value.
  • item: Conference-Full-text
    Statistical Machine Translation System for Sinhala and Tamil Languages
    Rajpirathap, S; Sheeyam, S; Umasuthan, K; Chelvarajah, A
    Statistical machine translation method is one of the most promising and efficient method to perform machine translation for Sri Lankan languages likes Sinhala and Tamil. Statistical approach is more suitable for structurally dissimilar pairs of languages and efficient solution for large text translation. In Sri Lanka we have a rising need for translation for Sinhala and Tamil and the statistical machine translation approach is more suitable for the concerned languages. Sinhala and Tamil have a similarity in grammar and statistical approach will help to obtain more accurate results. We have developed a bi-directional translation system for both Tamil to Sinhala and Sinhala to Tamil for this research. We have used the Sri Lankan parliament corpus to train the language model. We have critically evaluated the both systems with parameter optimizations and have obtained the most accurate and efficient system. We have also utilized the scoring techniques like BLEU [2, 8] & NIST [2] for the system evaluation and we have integrated the MERT technique to tune the decoder.
  • item: Conference-Full-text
    Sign Language to Audio Conversion for Sinhala Language with a Guided Framework
    Gunasekara, UDND; Dilhani, PAC; Thiripuvaneswaran, T; Ranathunga, L
    Communication is an essential part of all the people in their day today life. We can see various modes of communication with one another in our society. Ordinary people can communicate with each other using any language in the world if they are capable of that language. However, for the hearing impaired people communication is a difficult task as they are limited to the sign language. Many studies and applications have attempted to overcome this barrier with the usage of image processing and hardware technologies. Nevertheless there are very few solutions that support for the Sinhala language. Our system provides a solution for the above- mentioned matter by developing a system, which converts Sinhala sign language into Sinhala audio with the use of image processing technologies and a guided framework. We detect the hand gestures through a colored glove and the relevant word for a particular symbol can be retrieved from the XML file. The audio output is generated by using the festival framework. Our system has 80% accuracy for detecting Sinhala sign symbols.
  • item: Conference-Full-text
    Improving Human Performance in Robotics: An Augmented Reality Approach
    Lakshantha, E
    This paper presents our initial experimental work on monitoring human performance with a novel Augmented Reality framework. The aim of this novel framework is to use it for controlling robots in everyday environments such as in homes, schools and hospitals; hence improving the interaction performance of non-robotic experts. The paper explains the steps of our experiments in detail, including two case studies during which we applied our model framework to control robots in real world scenarios. Further details on the experimental outcome are provided in this paper, along with detailed results and analysis under each case study.
  • item: Conference-Full-text
    Biometric Signatures and Elliptic Curve Cryptography for Internet based Applications
    Perera, KMS; Weerasinghe, H
    Public key cryptography is an important aspect in Information Security. Since RSA is the most widely used and accepted public key algorithm so far, many proven applications based on the RSA algorithm are widely used around the world. At present, applications of the Elliptic Curve Cryptography have been arisen as ECC has become a competing algorithm to RSA. Therefore, many comparisons between the two algorithms have been identified for various applications. This research focuses on applying Public Key Cryptography for user authentication in online transactions. Authentication is carried out generating biometric base digital signatures. The process of generating the digital signature is a combination of biometric signature and public key cryptography. The introduced new technique was implemented for algorithms ECC and RSA, and then conducted a strength comparison between two algorithms in order to observe the most suitable algorithm for the biometric digital signature generation.
  • item: Conference-Full-text
    Agent Based Employee Performance Management and Decision Support System
    Ekanayake, DC; Gunasekara, RGAD; Weerasinghe, KHWMMM; Deemanthi, SHS; Sandanayake, TC; Fernando, S
    Performance appraisal is a process used by companies, in order to evaluate the employees efficiency and productivity. This process can be carried out in different ways, by employees supervisors or by different collectives related to the evaluated employee. When 360-degree performance appraisal approach is used to evaluate employees, different appraisers have different degree of knowledge about the employee. Such knowledge is usually vague, subjective and very limited. To overcome these limitations, this paper presents an agent based employee performance management and decision support system which enables an employee to represent himself in a virtual environment through an autonomous agent. This representation provides an opportunity for a person to justify the marks he/she has allocated for a given attribute to a virtual panel consists of his/her peers and managers. Furthermore, agents in this virtual environment are capable of acquiring the knowledge from other agents and update the beliefs according to feedbacks they receive and recalculate the evaluation marks.
  • item: Conference-Full-text
    Adaptive Surveillance System using a Camera Network
    Yatanwala, YWTM; Senadeera, KMSK; Ruhunuwickrama, OPI; Wijepala, NGNP; Ranathunga, L
    CCTV Surveillance Systems have been used in many places to address security issues. Those systems will store a large amount of activities in every moment and most of the times data is stored in separate places without having integrity. Considering large premises, more cameras and more operators are required to handle the situation. Continuous and careful attention is essential to identify the activities of a large crowd. Moreover, there is no proper mechanism to search and retrieve important information from the videos whenever it is needed. In this paper we introduce a new approach to uniquely address those issues, using different image processing algorithms and inter process communication techniques which were implemented to acquire more reliable and accurate timely information from CCTV records in advance.
  • item: Conference-Full-text
    A Portable Tool for Deaf and Hearing Impaired People
    Rupasinghe, RADK; Ailapperuma, DCR; De Silva, PMBNE; Siriwardana, AKG; Sudantha, BH
    A portable system for deaf and hearing impaired people has been developed based on the IEEE 1451 sensor standards and other related technologies to minimize their communication difficulties with the ordinary society. The system is comprised of two main modules. One of the modules is performing gesture recognition by capturing Sinhala sign language symbols from a deaf or hearing impaired person using a specially developed smart glove which is embedded with a 3-D accelerometer, a digital gyroscope and flex sensors complying with the IEEE Smart Sensor Standards. The smart glove is capable of capturing finger movements and hand movements. Arduino Mega controller system, containing Atmel ATmega2560 AVR microcontroller, was used to implement the Smart Transducer Interface Module (STIM). The captured signals were analyzed and related text and voice would be generated. The other module is processing human voice real-time using Sphinix-4 framework and generate corresponding text and displaying Sinhala sign language animations on an external display using an animated avatar. Generating animations was done by using Blender and JMonkey game engine.
  • item: Conference-Full-text
    A Novel Approach to Online Sinhala Handwriting Character Recognition on Handheld Devices
    Liyanaarachchi, VA; Jayasuriya, DB; Adikari, AS; Rodgrio, ST; Karunaratne, I
    Handwriting, being the main medium of written communication is now being overpowered by the digital technologies. There is a significant interest in addressing this issue and in facilitating handwriting recognition via electronic devices. Comprehensive studies can be found under this in both local and global contexts. In Sri Lanka, Sinhala language; being one of the official language, there are many approaches taken by the local science community to recognize offline and online Sinhala handwriting. However considering online Sinhala handwriting it lacks implementations for handheld devices. This study focuses on Online Sinhala Handwriting Recognition for handheld devices with the use of a Sinhala Spell Checker as a post processing technique. In this research, the writing angle, shape and size of the characters are used to recognize Sinhala handwriting. Therefore, the processing power and memory consumption is comparatively low thus making the solution efficient. Integration of a Sinhala spell Checker has increased the accuracy and handles any ambiguities in recognizing characters in a word. Sinhala Spell Checker is developed using n-gram probabilistic model and it is also capable of correcting errors occurred due to the phonetic similarity in letters. These techniques together provide a highly accurate solution for recognizing online Sinhala handwritten text, resulting an approximate accuracy of 89%.
  • item: Conference-Full-text
    A Comprehensive Approach for Mobile Shopping
    Sandakalum, MKDNP; Kulawansa, KADT
    One of the major concerns in mobile applications is the processing power and memory. In the meantime when an application uses a lot of processing and memory it tends to reduce the battery life of the device. This is a primary concern for developing advanced algorithmic functions to be able to run inhand held mobile devices. In order to cater to the requirements of the people, the mobile applications are needed in such a way where it consume only a little amount in the device, but capable of using the high tech hardware power runs on the server side environment. Mobile Shopper or ”mShopper” is an application which provides a solution to the problems in the area of retail shopping by adhering to the above mentioned concept. It runs the user interfaces in the mobile device consuming low processing power, but delegates the tedious processing elements including retrieving bulk data, multiagent processing and result formatting to a remote server. The application provides an easy mechanism to do the day to day shopping which uses mobile technology to manage the shopping activities in a distributed, and service oriented environment with the support of multiagent technology. The main focus of the application is to provide easy mobile interaction to the customers in their day to day shopping experience. The system manages all the complexity of the internal overhead including identifying locations, filtering lists, product searching, etc. and provides a convenient User Interface to use simple Input / Output in order to feel a shopping experience beyond traditional means.