International Research Symposium on Pure and Applied Sciences (IRSPAS)
Permanent URI for this communityhttp://repository.kln.ac.lk/handle/123456789/15650
Browse
2 results
Search Results
Item “MySight” – A Mobile based Solution for the blind community(Faculty of Science, University of Kelaniya, Sri Lanka, 2016) Nimalarathna, L.B.O.D.; Senanayake, S.H.D.Reading is an ultimate goal of human beings which is unfortunately not available for people affected by blindness. They can only read things which are in Braille system. But we all know that every book has not been converted to Braille system. The main problem of blind people for studying is that they cannot read and learn from textbooks. Therefore, it would be extremely useful to find a method, such that a blind person could read newspapers, textbooks, bills, etc. without the Braille system. Android platform can be used to enable this opportunity for blind people for the purpose of reading out loud any printed document or anything written using a standard font. This research is completely focused on finding a way which could let blind people be exposed to a new kind of reading. The main objective of this research was to develop an android application which could identify words and various kinds of symbols written using a standard font in a given document, and then convert them to an audible format such that a blind person could understand. It will be effective for a blind person by providing voice notifications and smart touch techniques. The first step was to find an appropriate and efficient Optical Character Recognition (OCR) technique which compatible with the Android platform. In order to fulfill that requirement, the Tesseract OCR library was used. After lying dormant for more than 10 years, Tesseract is now behind the leading commercial engines because of its accuracy. The next step was interface designing. In this application, there are two main interfaces. One is for capturing the text, and another one is for displaying the detected text. There are several operations that can be performed once it detects the text, such as, ‘read again’ function. Then, the OCR implementation for the Android platform has been successfully completed. For the future improvements, the application should be enhanced to guide the blind person to capture the image of the paper or the page of a book. For that, the OpenCV library can be used for paper detection and the smart voice commands for giving instructions to the blind.Item Incident crowdsourcing, tracking and ranking application for environmental problems and issues in Sri Lanka using natural language processing(Faculty of Science, University of Kelaniya, Sri Lanka, 2016) Lakshika, M.V.P.T.; Senanayake, S.H.D.Sri Lanka is having numerous critical environmental problems and issues such as deforestation, pollution of water bodies, natural disasters and many other urban problems. Many of the communities who suffer from such environmental problems are not attaining solutions for their vital problems due to lack of awareness, inefficiency and carelessness of amenable parties such as environmental related authorities and ministries in Sri Lanka. Within past few years, virtual communities in Sri Lanka used social media for emphasizing numerous forms of social problems. The intention of this research is to make the awareness of virtual community as a compulsion towards the responsible parties in Sri Lanka which can work as a driving force for stimulating reasonable solutions towards environmental problems. The web based application discussed in this research has been designed to obtain content of such environmental problems by soliciting contributions from crowdsourcing. Online community can report environmental problems by using text and images. Users of the application can vote and comment on the problems and issues posted in the application. Each problem will receive points based on up or down votes and comments they have received and then the application ranks genuine high quality environmental problems while allocating points for each user in the system. Text categorization is a subtask of information retrieval which used in this application is very effective for filtering of environmental related information before posting to the system. Further, this application is using the semantic information or the polarity of user comments as positive, negative or neutral which are not used yet for the most important natural language applications. This research discussed about a study of the interaction between Natural Language Processing and text categorization. Based on users negative or positive comments and up or down votes, the application calculates the points for each post according to a predefined criterion and highlights the genuine high quality environmental problems and issues in Sri Lanka.