https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/4740 Sun, 29 Mar 2026 07:51:21 GMT 2026-03-29T07:51:21Z https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/4810 Title: An algorithmic approach for automating the sorting and grading harvested TJC mangoes Authors: Patabendige, S.S.J; Sewwandi, A.V.P.; Tharaka, D.D. Abstract: Abstract Mango grading is a quality assessment method carried out by mango exporting industries. Precisely graded mangoes support to uplift the demand of the industry as well as the revenue. Currently, both manual and machinery-based approaches are being used in the global market. Sri Lanka follows a manual grading approach to grade TJC mangoes which is the predominant exporting mango variety in Sri Lanka with a beautiful golden orange and unblemished skin. However, manual grading of mangoes using visual perception is laborious, inaccurate, and inconsistent. Therefore, this project aims to overcome these issues by introducing a machine learning and computer-vision based solution to grade harvested TJC mangoes in terms of their surface spots, size, and weight. This research has focused on the six perspectives of each mango to get an accurate final result from the machine learning model for the mango grade. This work is carried out in four phases: The first phase is the acquisition of images for training and testing purposes. Images were acquired by capturing six images per mango from six perspectives. The generated dataset consists of 1500 color images from five different quality classes. For experimentation purposes, the dataset was split into two groups containing 1350 and 150 images respectively. The larger group was used to train the classifier whilst the smaller group was used as the test dataset. Several image pre-processing techniques were followed in the second phase to enhance the images and extracted the essential features for the classification processes to identify the color scale, thresholding, noise removal, and contour detection techniques. Morphological operations and histogram of gradient (HOG) and local binary pattern (LBP) were used in the feature extraction phase. K-means clustering and K-medoid was used to detect the surface around the mango stalk. Sequential forward selection algorithm was applied to identify the best feature subset from the extracted superset of features in the feature extraction phase. The final grade classifier considered all the six sides of images in each mango. In the fourth phase, a model consisting of two main classifiers for analyzing the size and surface spots was implemented. Fri, 26 Mar 2021 00:00:00 GMT https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/4810 2021-03-26T00:00:00Z https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/4809 Title: UrbanAgro : Application to Support Sri Lankan Urban Farmers to Detect and Control Common Diseases in Tomato Plants Authors: Fernando, W.L.V.; Navodi, H.A.D.D. Abstract: Abstract Plant diseases cause many significant damages and losses in crops around the world. Some appropriate measures should be introduced on identification of plant diseases to prevent damages and minimize losses. With Covid-19 lockdowns many Urban dwellers are encouraged to grow their own foods. As most urban farmers do not tend to use pesticides in their farms there is a high chance for the crops to get caught of various diseases. Comparatively, identifying the plant diseases visually is expensive, difficult and inefficient. And also getting expertise knowledge is very expensive and practically impossible to reach them whenever they need. As such this might be a difficult task for urban farmers or newcomers to this field to decide which disease can be attached to the crops. Early detection of diseases helps in increasing the productivity of crops as well as in minimizing expenses. Technical approaches using machine learning and computer vision are actively researched to achieve intelligence farming by early detection on plant diseases. The accuracy of object detection and recognition systems has been drastically improved by the recent development in Deep Neural Networks. The system proposed presents a practical, applicable solution for the identification of the type and location of 5 different types of diseased and healthy leaves of tomato plant, which is a significant difference from the conventional methods for plant disease classification. In this context we have used YOLOv3 model which is a method based on transfer learning to diagnose tomato plant diseases using images taken in-place by camera devices on smartphones instead of using the procedure to collect, test and analyze physical samples (leaves, plants) in the laboratory. The trained model achieved an average accuracy of 92 percent, which is exceptional in comparison to previous studies in this context. The target group of users are urban farmers who request a quick diagnosis on common tomato leaf diseases at any time of the day as they lack knowledge on diseases that are attached with plants. Fri, 26 Mar 2021 00:00:00 GMT https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/4809 2021-03-26T00:00:00Z https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/4808 Title: cGraph: Graph Based Extensible Cyber Threat Intelligence Platform Authors: Daluwatta, W. W.; De Silva, L. R. S.; Kariyawasam, S. N. Abstract: Abstract Cyber security, also referred to as computer security, is the field of study where computer systems and networks are protected from malicious digital attacks such as spam attacks, phishing attacks, malware attacks, and ransomware attacks from perpetrators. The number of cyber threats and malicious attacks continue to rise each year globally at a rapid pace and thus there is a need for implementing proper and effective cyber security measures and recognizing possible attacks or malicious internet resources early on. To address this issue, we propose ”cGraph: Graph Based Extensible Cyber Threat Intelligence Platform” which is a scalable big data processing and storing system that uses intelligence derived from state-of-the-art graph inference algorithms utilizing vast amounts of passive DNS network traces and limited amount of external threat intelligence. Fri, 26 Mar 2021 00:00:00 GMT https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/4808 2021-03-26T00:00:00Z