http://ir.lib.ruh.ac.lk/handle/iruor/16746 2026-04-28T09:10:22Z 2026-04-28T09:10:22Z Muhannadh, M.R. Laksiri, P.H.P.N. http://ir.lib.ruh.ac.lk/handle/iruor/16866 2024-04-17T10:12:54Z 2023-11-24T00:00:00Z

Dried fish type identification using machine learning techniques: A case study on three fish species in Sri Lanka Muhannadh, M.R.; Laksiri, P.H.P.N. Dried fish is one of the major and traditional dishes of Sri Lankans and is also one of the major animal protein sources with low cholesterol for humans. Despite, the benefits and the popularity, people are facing challenges in identifying certain varieties of dried fish which look the same such as Rastrelliger kanagurta (Kumbala), Goldstriped sardinella (Salaya) and Sardina pilchardus (Keeri). When the above three types of dried fish are mixed, the ability to differentiate one from another becomes challenging and benefits between the three varieties is a concern. Consumers have been misled by some sellers who manipulate the challenge of identification. Therefore, the aim of this research is to identify the correct type of dried fish even if those fish are mixed together with other types. To address the above issue, a machine learning based solution was developed along with image processing using more than 3500 images. The developed model uses four features of dried fish, the head, trunk, tail, and the entire body. Using YOLOv8 model, first we attempted to identify the number of objects in the user captured image and we store those identified objects temporarily within the model. Subsequently a trained VGG16 model is used for the classification of dried fish types based on the previously identified objects. The model achieved more than 90% overall accuracy in identifying the correct dried fish type. In conclusion, the developed model can be used to effectively identify the type of dried fish even in situations where they are mixed with other dried fish types. The model will be further developed as a mobile application in the future for the betterment of dried fish consumers as well as sellers.

2023-11-24T00:00:00Z Vidanapathirana, D.R. Arachchi, R.S.W. http://ir.lib.ruh.ac.lk/handle/iruor/16865 2024-04-17T10:04:22Z 2023-11-24T00:00:00Z

Oyster mushroom disease detection using machine learning Vidanapathirana, D.R.; Arachchi, R.S.W. The industry that cultivates mushrooms has experienced a significant expansion due to the growing demand for edible mushrooms and in particular for oyster mushrooms, as they are highly valued for their distinct flavor and high nutritional content. However, a major threat to the productivity is the susceptibility of oyster mushroom fields to diseases. This research presents a machine learning-based technique for the early detection of fungal infections in oyster mushrooms to address this issue. The main goals of the study are to create a large dataset of high-resolution images, provide a diagnosis method for several fungal diseases that affect oyster mushrooms, and carry out dataset splitting, augmentation, and preprocessing techniques. Using a dataset of 1500 data points, the study utilizes deep learning models and machine learning techniques like VGG16, ResNet50, and InceptionV3 to identify and classify oyster mushroom infections, demonstrating remarkable accuracy and precision in complex disease categories. This research significantly contributes to agriculture and the mushroom-growing industry, enhancing the understanding and classification of oyster mushroom infections. The study aims to use convolutional neural networks (CNNs) for feature extraction to create an accurate disease detection system for oyster mushroom fungal diseases. However, it acknowledges limitations like the need for a larger dataset and the need for diverse datasets for better generalization. Future research should focus on adding new characteristics to improve the accuracy of disease diagnosis. The proposed machine learning-based approach could revolutionize the mushroom cultivation industry by reducing financial losses from fungal infections and promoting greater yield sustainability. The research's potential benefits include early disease detection, prompt disease treatment, and reduced crop losses.

2023-11-24T00:00:00Z Karunasagara, D.K.M.U.K. Gamage, C.Y. http://ir.lib.ruh.ac.lk/handle/iruor/16864 2024-04-17T09:48:21Z 2023-11-24T00:00:00Z

The positive effect of food choices on academic stress among students at the University of Ruhuna Karunasagara, D.K.M.U.K.; Gamage, C.Y. Academic stress among university students is receiving widespread attention in today's academic environment. Academic stress arises from the myriad demands and challenges inherent in academic life, the management of which is fundamental to students' overall health and well-being. University students face a high sensitivity to academic stress and need proactive strategies to reduce stress. There is growing evidence that certain food choices can positively impact academic stress, but the exact link varies among individuals and depends on the specific foods consumed. This research project developed a prediction model using a machine-learning algorithm to determine the beneficial effects of dietary decisions on academic stress among students at the University of Ruhuna. The main goals are to determine the stress levels of the students, comprehend how they eat when under stress, and pinpoint foods that help reduce stress. The study combines supervised and unsupervised learning techniques using a two-pronged design. A dataset of 597 student participants, and a K-means algorithm are employed in the field of unsupervised learning to intelligently classify students into different stress levels based on their replies. This process revealed complicated patterns of food consumption. Simultaneously, supervised learning, facilitated by the K-Nearest Neighbors (KNN) algorithm, creates correlations between stress levels and personalized food consumption habits. The study concluded that there was a noteworthy pattern among students at University of Ruhuna with high stress levels, who consumed an average of 2.25–2.50 times more sweet foods than spicy and milky foods than their low-stress counterparts, who consumed an average of 1.00–1–25 times sweet foods. It illustrates the connection between a person's food intake and stress levels, as well as how eating well can temporarily reduce stress. These findings have important consequences for nutritional therapies that might be used to improve the ability to manage stress among students at University of Ruhuna.

2023-11-24T00:00:00Z Sandunima, W.L.G.D. Arachchi, R.S.W. http://ir.lib.ruh.ac.lk/handle/iruor/16860 2024-04-17T09:34:53Z 2023-11-24T00:00:00Z

Identification of teak wood cupboards in Sri Lanka using machine learning Sandunima, W.L.G.D.; Arachchi, R.S.W. Teak timber, scientifically referred to as Tectona grandis, is a timber species of exceptional value celebrated for its outstanding characteristics and wide-ranging uses. Teak wood cupboards hold significant value in the furniture industry, particularly in Sri Lanka, due to their durability, aesthetic appeal, and cultural significance. However, distinguishing authentic teak wood cupboards from imitations can be a challenging task, for both consumers and experts. This research presents a novel approach to address this issue by leveraging machine learning techniques for the automatic identification of teak wood cupboards. This study is confined to categorizing a collected dataset of 1060 cupboard images from furniture shops in Sri Lanka through image preprocessing. In this study, a machine learning model, specifically a Convolutional Neural Network (CNN), was developed and trained on a dataset of images of teak wood cupboards and other imitations of wood. The CNN model can recognize distinct features and patterns that differentiate genuine teak wood from other types of wood. The model's performance is evaluated, and the results indicate an accuracy of 89.5%, demonstrating its effectiveness in teak wood cupboard identification. To mitigate the limitations posed by a relatively small dataset, data augmentation technique was employed to prevent overfitting. Model performance was assessed using metrics like precision, recall, and the F-1 score. Additionally, it can contribute to the preservation of teak wood resources by discouraging the use of counterfeit materials on the market. The proposed model offers a promising solution to the problem of identifying teak wood cupboards in Sri Lanka, addressing both economic and environmental concerns.

2023-11-24T00:00:00Z