This thesis develops a comprehensive efficiency evaluation framework for 3D reconstruction from UAV imagery, addressing the critical need for objective algorithm selection and operational optimization. The research establishes systematic methodologies for evaluating computational efficiency, reconstruction quality, and cross-platform performance through multi-dimensional analysis rather than single-metric assessments. A novel Composite Unsupervised Efficiency Score (CUES) methodology enables multi-dimensional evaluation by integrating four distinct weighting approaches for objective algorithm assessment. The developed framework systematically evaluates 784 feature detection, description, and matching combinations across diverse datasets, providing evidence-based recommendations for algorithm selection, parameter optimization, and timing-based performance metrics. Statistical validation demonstrates consistent methodology reliability with high cross-dataset stability, strong ranking consistency between computing environments, and 95% confidence interval quantification enabling significance testing between algorithm alternatives. The framework includes enhanced COLMAP integration, interactive visualization tools, comprehensive performance databases, and standardized evaluation protocols. Key contributions include parameter optimization guidelines, cross-platform benchmarking across server and mobile architectures, multi-layered synthetic UAV datasets development, and novel efficiency metrics for operational UAV scenarios. This research establishes the first systematic efficiency evaluation framework for UAV-based 3D reconstruction, providing practical guidance for operational deployment optimization and setting new standards for objective performance assessment in computer vision applications.

In this thesis, currently available 3D mesh compression algorithms, frameworks, libraries etc. are investigated. Especially, the algorithms that are popular in survey papers but don’t have any implementation or had outdated implementation or no published version is available, are gathered together and compiled accordingly. According to the benchmark test results, current best general-purpose data compression methods are identified and applied as the last stage of mesh compression. Results are compared in order to demonstrate the current state of single-rate 3D mesh compression performance with the current best general-purpose data compression methods.

3D graphics are evolving media type used in all aspects of technological areas of today. Increase in demand on 3D graphics pushes technological advancements on 3D scan technology and approximation methods to next level which then results in more complex and highly detailed large 3D raw data. Thus, it is crucial to compress these graphics data efficiently. Over the last two decades, many algorithms have been proposed to compress these raw 3D data especially for compact storage, fast transmission, and efficient processing. Compression methods are branching among themselves. In this paper, 3D compression methods are summarized in a taxonomical fashion. A special attention is paid to the main ideas behind the single-rate compression algorithms and their contribution to 3D mesh compression technology. The advantages and the drawbacks of each algorithm are discussed to pave the road for the future 3D compression researchers.

Self-Organizing Maps (SOM), a type of Artificial Neural Network (ANN), is a data clustering tool that provides a way of representing multi-dimensional data in two-dimensional space. The maps are produced preserving topological relations between parameters of the input vectors. Unlike multi-layered feed forward neural networks, SOM employs unsupervised learning training mechanism. Interestingly, it requires no prior knowledge regarding the solution. The variety of the applications which employ SOM for data analysis reported in the literature is a clear indication of its acceptance as a powerful data analysis tool. SOM may, not only, present better viewing opportunities in such cases that displaying the relationships between the factors effecting the problem is impossible, but also, provides better exploration of the data. Application of SOM on the data collected in fisheries science provided enhanced outcomes and better understanding on the data collected. In this paper, SOM is discussed and reviewed in view of aquaculture and fisheries research based on the prevalence of isopods in the buccal cavity of one grouper species. The research was carried out to determine the seasonal patterns and potential impacts of the parasites on the goldblotch grouper using the SOM which were conducted in Iskenderun Bay.