Browsing by Author "Serugunda, Jonathan"

Now showing 1 - 5 of 5
  • Thumbnail Image
    Item
    A Deep Learning-based Detector for Brown Spot Disease in Passion Fruit Plant Leaves
    (arXiv preprint arXiv, 2020) Katumba, Andrew; Bomera, Moses; Mwikirize, Cosmas; Namulondo, Gorret; Ajeroy, Mary Gorret; Ramathaniy, Idd; Nakayima, Olivia; Nakabonge, Grace; Okello, Dorothy; Serugunda, Jonathan
    Pests and diseases pose a key challenge to passion fruit farmers across Uganda and East Africa in general. They lead to loss of investment as yields reduce and losses increases. As the majority of the farmers including passion fruit farmers, in the country are smallholder farmers from low-income households, they do not have sufficient information and means to combat these challenges. While, passion fruits have the potential to improve the well-being of these farmers given their short maturity period and high market value [1], without the required knowledge about the health of their crops, farmers can not intervene promptly to turn the situation around. For this work, we partnered with the Uganda National Crop Research Institute (NaCRRI) to develop a dataset of expertly labeled passion fruit plant leaves and fruits, both diseased and healthy. We made use of their extension service to collect images from five districts in Uganda to create the dataset. Using the dataset, we are applying state-of-the-art techniques in machine learning, specifically deep learning at scale for object detection and classification for accurate plant health status prediction. While deep learning techniques have been applied to various disease diagnosis contexts with varying degrees of success([2], [3], [4], [5], [6]), there has not been any significant effort, to the best of our knowledge, to create a dataset or apply machine learning techniques to passion fruits despite their obvious financial benefits. With this work, we hope to fill this gap by generating and making publically available an image dataset focusing on passion fruit plant diseases and pest damage and training the first generation of machine learning-based models for passion fruit plant disease identification using this dataset. The initial focus is on the locally prevalent woodiness (viral) and brown spot (fungal) diseases.
  • Thumbnail Image
    Item
    Face Recognition as a Method of Authentication in a Web-Based System
    (arXiv preprint arXiv, 2021) Mugalu, Ben Wycliff; Wamala, Rodrick Calvin; Serugunda, Jonathan; Katumba, Andrew
    Online information systems currently heavily rely on the username and password traditional method for protecting information and controlling access. With the advancement in biometric technology and popularity of fields like AI and Machine Learning, biometric security is becoming increasingly popular because of the usability advantage. This paper reports how machine learning based face recognition can be integrated into a web-based system as a method of authentication to reap the benefits of improved usability. This paper includes a comparison of combinations of detection and classification algorithms with FaceNet for face recognition. The results show that a combination of MTCNN for detection, Facenet for generating embeddings, and LinearSVC for classification out performs other combinations with a 95% accuracy. The resulting classifier is integrated into the web-based system and used for authenticating users.
  • Thumbnail Image
    Item
    Machine Learning-Aided Optical Performance Monitoring Techniques: A Review
    (Frontiers in Communications and Networks, 2022) Tizikara, Dativa K.; Serugunda, Jonathan; Katumba, Andrew
    Future communication systems are faced with increased demand for high capacity, dynamic bandwidth, reliability and heterogeneous traffic. To meet these requirements, networks have become more complex and thus require new design methods and monitoring techniques, as they evolve towards becoming autonomous. Machine learning has come to the forefront in recent years as a promising technology to aid in this evolution. Optical fiber communications can already provide the high capacity required for most applications, however, there is a need for increased scalability and adaptability to changing user demands and link conditions. Accurate performance monitoring is an integral part of this transformation. In this paper, we review optical performance monitoring techniques where machine learning algorithms have been applied. Moreover, since many performance monitoring approaches in the optical domain depend on knowledge of the signal type, we also review work for modulation format recognition and bitrate identification. We additionally briefly introduce a neuromorphic approach as an emerging technique that has only recently been applied to this domain.
  • Thumbnail Image
    Item
    Network Reliability Analysis as a Tool to Guide Investment Decisions in Distributed Generation
    (Journal of Power and Energy Engineering, 2018) Ssemakalu, Samson Ttondo; Edimu, Milton; Serugunda, Jonathan; Kabanda, Patrick
    Distributed Generation (DG) in any quantity is relevant to supplement the available energy capacity based on various locations, that is, whether a site specific or non-site specific energy technology. The evacuation infrastructure that delivers power to the distribution grid is designed with appropriate capacity in terms of size and length. The evacuation lines and distribution network however behave differently as they possess inherent characteristics and are exposed to varying external conditions. It is thus feasible to expect that these networks behave stochastically due to fault conditions and variable loads that destabilize the system. This in essence impacts on the availability of the evacuation infrastructure and consequently on the amount of energy delivered to the grid from the DG stations. Reliability of the evacuation point of a DG is however not a common or prioritized criteria in the decision process that guides investment in DG. This paper reviews a planned solar based DG plant in Uganda. Over the last couple of years, Uganda has seen a significant increase in the penetration levels of DG. With a network that is predominantly radial and experiences low reliability levels, one would thus expect reliability analysis to feature significantly in the assessment of the proposed DG plants. This is however not the case. This paper, uses reliability analysis to assess the impact of different evacuation options of the proposed DG plant on its dispatch levels. The evacuation options were selected based on infrastructure options in other locations with similar solar irradiances as the planned DG location. Outage data were collected and analyzed using the chi square method. It was found to be variable and fitting to different Probability Distribution Functions (PDF). Using stochastic methods, a model that incorporates the PDFs was developed to compute the reliability indices. These were assessed using chi square and found to be variable and fitting different PDFs as well. The viability of the project is reviewed based on Energy Not Supplied (ENS) and the anticipated project payback periods for any considered evacuation line. The results of the study are also reviewed for the benefit of other stakeholders like the customers, the utility and the regulatory body.
  • Thumbnail Image
    Item
    Optimizing Location of Edge Clouds with Baseband Units in Cloud Radio Access Network
    (IEEE., 2019) Nakazibwe, Jackline; Serugunda, Jonathan; Akol, Roseline; Mwanje, Stephen
    Cloud radio access network architecture has become increasingly important in meeting the growing demand of high data rate services as well as managing interference among base station sites. However, this architecture is associated with high fronthaul link latencies that result into increased overall network latencies. This paper proposes a placement method for baseband units so as to lower the fronthaul link latencies. This is achieved by formulating the problem as a nonlinear optimization problem that is solved by fuzzy c-means clustering and a heuristic genetic based algorithm. The paper further investigates the user response time comparing it with the scenario before optimizing location of baseband units in edge clouds. Simulation results show that the proposed scheme greatly reduces overall fronthaul link latencies and cost of ownership in cloud radio access networks. We also show that there is an optimal number of edge clouds with baseband units for different cloud radio access network sizes. On the side of the users, the response time is greatly reduced when the baseband units are optimally placed.