https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/4556 2026-04-05T23:21:20Z https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/3258 Title: Remotely Sensed Image Classification Technique for Tea Plantation in Sri Lanka Authors: Mahavidanage, M.D.S.R.A. Abstract: In Geographic Information Systems, supervised learning is traditionally used to classify remotely sensed imagery data to develop land-use maps. The classifiers used for this process tend to generate inconsequent classes due to the variety of reasons. To understand the problem, several experiments were done by using a number of commonly used unsupervised and supervised image classifiers namely K-means clustering, ISODATA clustering, Maximum Likelihood, Mahalanobis Distance, Minimum Distance, Parallelepiped and Binary Encoding. These experiments were done using the ERDAS Imagine and ENVI image processing software applications. The sample image chosen is a subset of Quick Bird-2008 Satellite imagery of Ganga Ihala Korale divisionin Kandy district in Sri Lanka covering an extent of 5.73ha. None of the approaches experimented generated the expected land-use maps for the tea plantation. Although Mahalanobis Distance classifier achieves the highest accuracy in confusion matrix, the result is not suitable to use for a land-use map as it generates a number ofinconsistent and compound classes.Thus, this study is aimed at exploring a new approach to identify the tea plantation in Sri Lanka for land-use mapping while reducing the misclassification errors generated by the other approaches. The proposed approach is based on the supervised classification approach which uses the conventional minimum distance decision rule. It incorporates a spatial object based threshold scheme to reduce the complexity of the classification and to reduce the effect of the misclassification problem. The proposed method is tested for upcountry tea plantation of Sri Lanka. Two subsets of Quick Bird-2008 Satellite imagery in Kandy district are used to evaluate the proposed approach. The results show that the pixels are correctly classified to the correct classes in the GIS layer which consists of spatial boundaries of land-use types. The comparison of the results of proposed method with the conventional method reveals higher accuracy in proposed method. 0019-11-15T00:00:00Z https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/3256 Title: Holistic Approach in Recognizing Handwritten Tamil words Authors: Thadchanamoorthy, S. Abstract: Optical Character Recognition, OCR, is the process of converting the images of handwritten, typewritten, or printed text into machine editable text, such as ASCII code. The area of OCR concerns the essential concepts of pattern recognition. The handwriting recognition can be seen as a sub task of OCR. It provides simple interface between man and machine. In spite of several advancements in technologies pertaining to optical character recognition, handwriting continues to persist as means of documenting data for day to day life. In the field of handwriting recognition, on-line and off-line recognitions are traditional. The handwritten character recognition is a very difficult process due to the cursive and unconstrained nature of the handwritten characters due to the different styles of different writers. Further, the handwritten characters are sometimes overlapped and touched with the adjacent characters. This is one of the major hurdles in segmentation of characters from the words. Therefore, nowadays a holistic approach together with other techniques becomes popular in recognizing handwritten words, rather than recognizing individual characters. The holistic word recognition approach is mainly used in the area of postal automation, bank check processing, automatic data entry etc. In this research, due to the cursive style of handwritten Tamil scripts, two classification models using holistic approach for handwritten Tamil words are proposed. The first model is based on simple geometric features using SVM classifier and the second model is based on the directional features using MQDF classifier. The importance of the first approach is the improvement on input images prior to the feature extraction. In addition to the generally available prepossessing techniques such as Otsu’s binarization, standardization, thinning process, and the slant correction, some additional corrective measures such as removal of unwanted prolongs (pruning and clipping) and mid alignment of the characters are proposed to improve the word images. A dataset is created including 218 country names, 156 Sri Lankan city names and 109 Tamil Nadu city names for the purpose of this research. For each of the name, one hundred samples are collected from a group of 500 different writers. The first approach is based on geometric features using Gabor filter. The country names (217) were considered for this work. They are the number of vertical lines, the number of v horizontal lines, number of +45 degree slanted lines, number of -45 degree slanted lines and the number of dots appeared in the word image. Further, these features are counted at twelve different positions on a 3x4 gridded word so that to increase the intra word variations. A simple technique to compensate the loss of vertical number of lines due to touching (within and between characters) is also proposed. A significant result with accuracy of 86.36% is achieved. The second attempt is targeted for the postal automation in Tamil Language within Sri Lanka and Tamil Nadu, India. For this purpose, the famous split-and –merge algorithm is used. Avoiding proper segmentation, a city name string is considered as a word and the recognition problem is treated as lexicon driven word recognition. In this approach, binarized city names are pre-segmented into primitives (individual character or its parts) using water reservoir concept. Primitive components of each city name are then merged into possible characters to get the best city name using dynamic programming. For merging, the total likelihood of characters is used as the objective function and character likelihood is computed based on Modified Quadratic Discriminant Function (MQDF), where four directional features (horizontal, vertical, 45 degree slanted and 135 degree slanted) are applied. From the experiment, a significant result of 96.89% accuracy is obtained, out of 265 word classes 0019-11-15T00:00:00Z https://dl.ucsc.cmb.ac.lk/jspui/handle/123456789/3255 Title: A thesis submitted for the Degree of Master of Philosophy Authors: Thilakarathna, P. K. M. Abstract: I observed that there is a spontaneous informal peer group activity named “Kuppi” among Sri Lankan university students. Inspiration to study about Kuppi in more detail lead me to conduct surveys and interviews with undergraduate students at University of Colombo School of Computing. Thereby, I found out that Kuppi is an informal reciprocal teaching activity and very popular among the undergraduate students. I observed that students conduct their Kuppi sessions often at locations outside the institute i.e. for example cafeteria. Since it is difficult to acquire conventional mediation tools and artifacts such as white boards to support learning at an outdoor location, conducting group learning activities in such resource constrained environments is tedious. Nowadays, most students carry wireless connectivity enabled laptop computers which can successfully be used as a mediation artifact in Kuppi. However, results of the studies also revealed that even if they have several laptop computers available, they only use single laptop computer for Kuppi as they find it difficult to use several laptop computers interactively within a group discussion conducted outside of institutional settings. A comprehensive study of existing potential tools that support collaboration and group learning activities revealed that none of them address the specific requirements of Kuppi. Therefore, my main objective in this research was to develop an appropriate framework to help students to conduct Kuppis using multiple laptop computers. I designed an application named “KuPpI” considering the requirements gathered during initial phase of my research. I kept the design as simple as possible to make sure students don’t have to spend much time on setting it up KuPpI prior to a Kuppi session. KuPpI was tested and evaluated for its usability revealing good results in performance and user interface design. KuPpI was then given to student groups for evaluation by using it in their Kuppi sessions. Observations, focus group discussions and surveys were carried out to evaluate such events and get students’ feedback. Analyzing collected data, it was evident that students find KuPpI very useful for their Kuppi. Furthermore, results revealed that, using KuPpI, they can easily connect multiple laptop computers in a location where no existing network infrastructure available and using KuPpI features such as ability to share screens; sending messages to group members within a Kuppi session; and transferring files interaction has become easier. Thereby, problems they encountered in using laptop computers when conducting Kuppi have been successfully addressed by KuPpI. Removing the environmental restrictions of a learning activity improves the effectiveness of the activity. Through KuPpI, I have successfully found a way to improve the effectiveness of Kuppi. 0019-11-15T00:00:00Z