During search and rescue missions, the rescue team must look for potential victims after a disaster has taken place and bring them to safety as quickly as possible. Lack of visibility of the victims and their location can waste time and resources and significantly hinder this procedure. In this project, we explore the theory behind deep learning, computer vision specifically object detection, and develop deep learning methods for object detection from aerial infrared (FLIR) thermal images taken by a UAV, with applications in the development of navigational aids for search and rescue operations. We demonstrate that a Faster R-CNN object detection network can be trained and fine-tuned to detect specific objects (survivors) in aerial images. We also seek to demonstrate that using multiple datasets in combination during the training of the network can provide significant performance improvements on test data.

This study examined the major determinants of savings in Uganda using secondary annual time series data for the period 1987-2018. The macroeconomic variables used include GDP, gross domestic product per capita GDPC, final consumption expenditure FCE, Gross national expenditure, inflation, merchandise trade, tax on income profits and capital gains and broad money supply. The model was estimated using co-integration and error correction model to analyze the short and long run equilibrium among the variables. The ADF test shows that gross domestic savings, merchandise trade and tax on income profit and capital gains contained unit root at levels. However, all variables were stationary after first difference. Results of the study showed that most variables except Gross national expenditure and broad money supply play a significant role in determining the savings rate in Uganda. The coefficient analysis also shows that GDP, gross domestic product per capita, merchandise trade and broad money supply have a positive impact while the final consumption expenditure, gross national expenditure and inflation revealed negative impact on the savings rate in Uganda. The study also concludes that the speed of adjustment is 18.6% which means that the model will correct short run disequilibrium position at the rate of 18.6% annually. The paper concludes that there is a potential for increased domestic savings and this calls for policy changes and political will among policy makers who should put more emphasis on final consumption expenditure, inflation, taxes, gross domestic product per capita, taxes on incomes profits and capital gains and merchandise trade that influences saving.

In this project, it was proposed to carry out research that involves the development of a central- ized Equipment Scheduling System Using AI-Based Authentication for researchers and research institutions in Uganda, based on the microservices system architecture. The system develop- ment involved the usage of the agile method of software development to achieve continuous delivery and integration. Simultaneously, a face recognition model was trained using a com- bination of data collection, localization, transfer learning and classication of processed data. The model was then integrated into the system as a service. The system allows users to login with a face recognition option, and browse through the cat- alogue of available resources, enabling them to request access to use special equipment or a resource that is owned by other institutions after which the system schedules the request.

This project was aimed at determining the effects of wood ash admixture on the engineering properties of burnt clay bricks. Sieve analysis and Atterberg’s limits tests were carried out to classify the clay soil used. These tests indicated that it was poorly graded clay soil of intermediate upper plasticity. The clay soil had a liquid limit of 45.1%, Plastic Limit of 13.4% and Plastic Index of 31.7 %. Bricks were made with partial replacement of clay using wood ash in percentages of 0%, 5%, 10%, 15% and 20%. These bricks were tested for water absorption, dry compressive strength, wet compressive strength and density. Density was observed to gradually decrease with increase in wood ash content. Maximum density was observed at 1833 kg/m3. A maximum dry compressive strength of 14.3 N/mm2 and maximum wet compressive strength of 9.7 N/mm2 were observed at 5% clay soil replacement with wood ash. The lowest water absorption of 14.78% and highest softening coefficient of 0.68 were also observed for bricks of 5% wood ash content. Bricks made with 5% wood ash content had the highest dry and wet compressive strength, lowest water absorption, highest softening coefficient and were less dense than the control (normal) bricks. We recommend that these bricks be adopted for use in the construction industry since they have improved engineering properties as compared to the normal bricks. Using these bricks can help reduce damages that occur due to breakages during transportation, handling and construction.

The main aim of the study was to investigate the spatial distribution and usability of KCCA public toilets in Central division. The specific objectives for this study included; • To investigate the spatial distribution of KCCA public toilets and how it influences their usability in Central division of Kampala • To investigate the challenges faced by the management and the users of the KCCA public toilets • To suggest strategies that would optimize the usability of KCCA public toilets in central division of Kampala Various methods and techniques of data collection were used to collect the data and these included, observation, photography, mapping, questionnaire, interview. Analysis of the data collected was done inform of maps, tables, pie charts and descriptive analysis. In this study, analysis shows that the KCCA public toilets are unevenly distributed and are concentrated in the civic area and Nakasero areas within the central division of Kampala. The study revealed a total of nine KCCA toilets within central division of Kampala and each of them has two cleaners to cater for both sex cubicles. Additionally, the users of these toilets prefer to use these KCCA public toilets because they are free of charge and additionally most of these toilets favor the pedestrians as they lack packing space for people who use vehicles. The study also analyzed the challenges faced by the management and users of the KCCA public toilets. Some of the key research findings reveal there is limited information about the location of these toilets, absence of showering facilities for the users, lack of parking space for users like the boda boda men. Furthermore, other challenges include politically driven decisions which influence the location of these toilets. Public toilets are very important for any country as their usability is dependent upon their spatial distribution and therefore, they ought to be designed to cater for all users and provide adequate facilities.