Browsing by Author "Wickramarathne, S.D.H.S."

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    Developing Simple and Economical Prototype to Measure the Internal and External Quality Parameters On Poultry Eggs
    (Uva Wellassa University of Sri Lanka, 2019-02) Ranchagoda, B.S.; Udayangani, J.T.C.; Sandaruwan, T.H.S.S.; Wickramarathne, S.D.H.S.; Abeyrathne, E.D.N.S.
    Egg quality is based on the characters of an egg that affect its acceptability to the consumer. There are several quality parameters used to identify the quality eggs; both internal and external. Internal quality refers to measuring the egg albumin and yolk distribution and yolk color whereas external quality refers to egg size, weight, shape and shell thickness. According to the quantitative and qualitative parameters there are different grading systems available for measuring quality and classify eggs with different groups and price levels. However, still there is no user friendly and economical method to measure the egg yolk color, shape, weight and shell thickness from a single machine. The study attempts to design a prototype to fill this gap. The egg weight is measured by using load cell which is connected to a liquid crystal display and to the computer. When the egg is placed on the cell the value is displayed both on the computer screen and display. The shell thickness is measured using a digital Vernier caliper. The value is displayed once it is placed between the two arms of the Vernier caliper, which is connected to the computer via an Arduino board. The yolk color and egg shape are measured using image processing techniques. In both processes an RGB image is taken and it is converted to a gray scale image. Then a histogram is developed using the pixel count of each point through length and width. Finally, by analyzing the histogram the output is given. The completed prototype was tested and accuracy was measured. Each feature of the model was accepted by more than 60% of accuracy. After all, a survey was done for testing the user acceptance with the participation of selected 20 poultry farmers, and that accuracy level was appreciated by more than 80% of respondents.
  • No Thumbnail Available
    Item
    Recommender System Based on Food and Exercise Ontologies to Find the Suitable Fitness Exercise Plan with the aid of Python
    (Uva Wellassa University of Sri Lanka, 2021) Basnayake, P.B.M.C.S.; Peiris, H.C.S.; Wickramarathne, S.D.H.S.; Jayathunga, D.P.
    In the modern world, professionals of diverse industrial sectors have severely become victims of overweight and obese conditions which can be minimized by having proper dietary plans, physical activities, and minimizing alcohol-based relaxation. However, most of the exercise plans provided by fitness applications currently available for usage are not personalized and general exercises are given for every individual. In this research context, individuals are guided by recommending suitable exercises with exercise frequency, exercise environment, and unique time period to perform according to body parameters. According to domain experts, fitness plans highly depend on individual characteristics. Therefore height, weight, age, sex, diet details, medical history and user preferences for exercises taken from the front end which is a Tkinter Graphical User Interface. In this system, food ontology uses these details to calculate the daily calorie intake and extra calorie intake of the particular individual. Disease extraction using natural language processing techniques, computed with Python and integrated with the output of Food Ontology which is to be mapped with the exercise ontological knowledge base along with the predefined rules to match respective exercises suitable for the particular individual that is compatible with his preferences. Two ontologies for foods and exercises developed using Protégé 4.3 and data retrieved by running Simple Protocol and Resource Description Framework Query Language (SPARQL) queries inside the Python code using the RDFLib module and output is taken and directed to the front end. The entire system developed with Python 3, where two ontological files of Foods and Exercises are loaded and tested for consistency using the HermiT reasoner with the aid of Owlready2. The task-based ontology evaluation approach is performed by addressing the competency questions through the execution of SPARQL queries. In conclusion, this study provides an approach to integrate two ontologies and a disease extraction model using Python programming language. Correctness and qualitative evaluations of the system are verified by the domain experts, and recommendations from the ontological system are beneficial for physical trainers to improve and validate their manual exercise recommendations. Keywords: Exercises; Ontology; Food; Tkinter; Python
  • No Thumbnail Available
    Item
    Smart Dustbin with a Web Based Point Reward System for Waste Management
    (Uva Wellassa University of Sri Lanka, 2018) Maduranga, G.K.C.D.; Gayanthika, W.A.L.; Silva, A.I.S.; Wickramarathne, S.D.H.S.; Ranasinghe, R.M.I.S.
    Sri Lanka is facing urbanization with the impact of globalization, which has results in an increase of needs and wants of people and increasing living standards of people tremendously. This economic and social advancement has resulted in a large amount of waste production in the urban areas of the country. The Government is still struggling to find a mechanism to manage this vast waste amount. The main issue government faces when managing waste is collecting pre-categorized garbage at collection points. The main reason behind this issue is lack of motivation of people to put garbage in to correct dustbins. The lack of mechanism to make aware the garbage collection center when the dustbins are full is another problem in current waste collection process. The researchers propose an Arduino based smart dustbin to overcome the above mentioned issues. The smart dustbin is integrated with a RFID reader which can identify its users and opens its lid only for the authorized users. A set of smart dustbins are assigned to certain number of households depending on their waste amount. When the user dump garbage in to the dustbin, the dustbin will measure the weight of the garbage using a load-cell and the user will be rewarded with eco points if a user dumps according to the relevant categories of garbage. Users can convert these eco points to financially benefited offers. The user can view their point balance and their details by accessing their profile through a web site. When the dustbin reaches its overflow level it is identified using an ultrasonic sensor and using a GSM module the relevant authorities are notified with a text message that the dustbin is full and about to overflow. These notifications help authorities to make the waste collection process smooth and effective. Eco points collection mechanism motivate people to dispose garbage in to categories and the smart dustbin make the waste collection process more effective and smooth. If the government involve in replacing conventional dustbin with this smart dustbin, this system will be an environmentally acceptable and economically feasible solution for the disposal of solid waste in Sri Lanka.
  • No Thumbnail Available
    Item
    Sri Lankan Sign Language Tutor
    (Uva Wellassa University of Sri Lanka, 2013) Fernando, K.S.S.; Wickramarathne, S.D.H.S.
    Sign Language Recognition is one of the major research areas of Human Computer Interaction (HCI). A large number of researches had been done in this area for American Sign Language, Indian sign language, Chinese sign language, Thai sign language, etc. (Rajathi et al., 2013). But, less amount of researches are done related to Sinhala sign language recognition. Specially no research work found in developing a tutor for Sinhala sign language. There may have problems when teaching sign language to disabled children due to lack of teachers, less/no attention to every child at every moment due to lack of resources, parents of these disabled children may be too busy, less interest of children to study, etc. Therefore, this research was carried out to develop an automated tutor to the Sri Lankan deaf community for Sinhala sign language to practice & check their knowledge. Methodology This is an Image based Sign Recognition System which uses Fourier Descriptors for feature Extraction (Nixon et al., 2002). The system architecture can be divided into five modules as shown in Figure 1. In Image Acquisition Module, images of 200*200dpi (dots per pixel) resolution which contains only the hand was captured in a black background by wearing a black long sleeve top with a white glove. In Image Processing Module, image processing techniques will be applied to manipulate the image. It results the pure contour of the shape as the final output. In Feature Extraction Module, Fourier Transformation is applied on closed contour. For that the shape contour is sampled to 64 points by equal arc length method and Centroid distance function is applied. Discrete Fourier Transformation is applied to the Centroid distance shape signature which result 32 Fourier Descriptors. The Fourier Descriptors were modified in order to preserve scale and rotation invariance by dividing them by the first Fourier Descriptor. The first 10 Fourier Descriptors except the first Fourier Descriptor are taken into account. In Neural Network Module, a neural network of 10 input layers, 6 output layers and 8 output layers is used. The 10 Fourier Descriptors of each sign are fed as the input and a pattern for each sign was fed as the output. It will output the results as a neuron weight file. This weight file is used in Result Module to determine the correctness (right/wrong) of each sign to the user. The result will be displayed to the user by a two visual indicators. Result and Discussion Sri Lankan Sign Language Tutor was implemented to recognize the correctness of 8 static signs of Sinhala Alphabet (Figure 2). The system is tested with 800 images of signs including 100 images from each sign for scale, rotation, translation and starting point invariance and the obtained accuracy level for each letter is listed in Table 1. The system is trained by using 200 images of signs including 25 images from each sign. The obtained accuracy level can be increased by increasing the training set. But it requires more time.