http://dlib.iit.ac.lk/xmlui/handle/123456789/2297 2026-04-06T22:10:34Z 2026-04-06T22:10:34Z Baskaran, Rushanthi Fernando, Pumudu http://dlib.iit.ac.lk/xmlui/handle/123456789/2296 2025-05-01T03:46:28Z 2021-01-01T00:00:00Z

Steel Frame Structure Defect Detection Using Image Processing and Artificial Intelligence Baskaran, Rushanthi; Fernando, Pumudu Steel Frame Structure Defect Detection is one of the main stages in constructing a building, where most of the time it has been done manually, which leads to no proper inspection. The aim of this paper is to detect six main defects in welded steel frame structure by using image processing and deep learning algorithms, where the application would aid individuals in construction sites to identify defects in said steel frame structures at an early stage of building in order to avoid casualties caused by the defect. An android application incorporated with a classification model was proposed and built. In this research, MobileNet has been used as the classifier algorithm, where Transfer Learning has been implemented on the pretrained model on ImageNet. CNN layers have been customized where GlobalAveragePooling2D layer has been implemented with Rectified Linear Unit being the activation layer and being fed into SoftMax layer. Furthermore, SGD optimizer with Categorical Cross Entropy Loss functionality have been applied. An image preprocessing of data augmentation and image transformation have been done. Viewpoint range is achieved at 10 – 3cm and error free under device rotation circumstances. Robustness and processing performance of the application have been achieved to an optimum level since it runs locally. The mean accuracy level of the device has been achieved for 91% for scratch, 78% for patches, 81% pitted surface, 78% for crazing, 73% for rolled in scale and 67% for inclusion defects in welded steel frame structure which sums up with a model mean accuracy being at 78%.

2021-01-01T00:00:00Z Vimaleswaran, Brindahini Ratnayake, Gayashini Shyanka http://dlib.iit.ac.lk/xmlui/handle/123456789/2293 2025-05-01T04:10:03Z 2021-01-01T00:00:00Z

E - Therapy Improvement Monitoring Platform for Depression using Facial Emotion Detection of Youth Vimaleswaran, Brindahini; Ratnayake, Gayashini Shyanka In today's generation depression is the most common mental health disorder which is mainly affecting many people's lifestyles especially the youth generation. There are lot of applications evolved for depression and recent review of depression apps indicates that importance given for video communication in online therapy is low which leads to difficulties while therapist and patients are engaging for therapies. It is also reported that these applications lack proper evidence of improvements which is a significant concern of the therapist and the users in order to monitor the therapy progress. Hence the ultimate aim is to provide an effective application for online depression therapy which has the ability to predict depression of the patient, assist the therapist and the patient to monitor the patient's improvement level and progress on each session using facial expression recognition. The proposed solution consists of depression prediction using feed - forward neural network model and Depression scale to measure the depression level. Calculating the improvement level is based on the depression levels identified and finally Visual represented in a dashboard to monitor depression level improvements for the therapist and the patient.

2021-01-01T00:00:00Z Peduru Hewa, Duneesha Suloshini Farook, Cassim http://dlib.iit.ac.lk/xmlui/handle/123456789/2292 2025-05-01T04:51:30Z 2021-01-01T00:00:00Z

A Sinhala Natural Language Interface for Querying Databases Using Natural Language Processing Peduru Hewa, Duneesha Suloshini; Farook, Cassim In this paper, the author presents SinSQLFinder, a Sinhala natural language user interface for generating SQL queries. Data is the heart of the decision-making process in every business, every organization, every government office. But at the same time working with data stored in the databases require special technical skills like Structured Query Language (SQL). It is an identical problem that the non-technical people are facing difficulties while retrieving data from the databases. In Sri Lanka, the native language of the majority of people is Sinhala and a considerable amount of e-governance applications use relational databases. Therefore, to manipulate data from such database applications easily, non-technical users who are more confident with the Sinhala language, need a solution to agree with a simple sentence in Sinhala and generate a valid SQL query. Nevertheless, the main goal of this research has been to establish a strong link between the Sinhala language and structured query language. Because of the language’s complexity, a unique method was developed to translate Sinhala language questions into structured query languages

2021-01-01T00:00:00Z Liyanage, Oshadhi Jayakumar, Krishnakripa http://dlib.iit.ac.lk/xmlui/handle/123456789/2291 2025-05-01T12:18:32Z 2021-01-01T00:00:00Z

Hate Speech Detection in Sinhala-English Code-Mixed Language Liyanage, Oshadhi; Jayakumar, Krishnakripa With the steady increase of user-generated content on the internet, the amount of hate content on the internet is also being rapidly increased. Social media sites, review forums, microblogging sites encourage users to convey their thoughts with minimum restrictions. This leads to expressing hate towards others who do not believe their beliefs. This study focuses on identifying hate speech texts that are written in Sinhala-English code-mixed language (Singlish) which is mostly used by Sri Lankans on the internet. Due to the unavailability of Sinhala-English code-mixed datasets, the dataset was created using comments on YouTube and Facebook. In this research, eight machine learning algorithms and three ensemble approaches were evaluated to detect hate speech in Singlish. Furthermore, their accuracy, precision, recall, and f1-score were evaluated. Afterwards, based on the performance of the considered algorithms, Support Vector Machine (SVM), Multinominal Naïve Bayes (MNB), AdaBoost Classifier, and Logistic Regression classifiers were used to develop ensemble learning-based solutions. In terms of ensemble learning approaches, soft voting, hard voting, and stacking were evaluated. The hard voting approach outperformed other baseline algorithms and ensemble approaches with 84% accuracy and f1-score.

2021-01-01T00:00:00Z