Browsing by Author "Ally, Juma Said"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Enhancing coffee leaf rust detection using DenseNet201: A comprehensive analysis of the Mbozi and public datasets in Songwe, Tanzania
    (African Journal of Empirical Research, 2025) Karia, Adrian Jackob; Ally, Juma Said; Leonard, Stanley
    Coffee Leaf Rust (CLR) is a worldwide devastating fungal disease that threatens coffee production, upsetting economic and farmers' livelihoods. Traditional methods of detecting CLR heavily rely on using machine-learning (ML) models trained through weakly collected datasets and physical inspection, which is tedious, time-consuming, and subject to human error. This study explores the performance of the DenseNet201 model using three datasets: Mbozi, Public, and Combined (a merger of Mbozi and Public datasets). Machine Learning Theory guided this research. The study objective is to assess the influence of dataset quality on CLR detection, analyze Mbozi and Public datasets using DenseNet201, and enhance robustness by merging the two datasets. A study of CLR severity was conducted using systematic sampling techniques. Leaves from multiple coffee farms were collected, representing different levels of infection. The Mbozi dataset, sourced from high-resolution images captured from Tanzania's Songwe coffee plantations, was analyzed for quality under controlled conditions, including environmental factors, image clarity, resolution, labeling consistency, and class balance, based on data completeness, image quality score, visual inspection, and model performance. DenseNet201 was trained and validated on each dataset, achieving its highest accuracy with the Mbozi dataset at 98.72% and a validation accuracy of 97.65%, demonstrating the importance of consistent image quality and accurate annotations. In contrast, the public dataset suffered from inconsistencies in resolution and labeling, resulting in a lower training and validation accuracy of 96.86% and 96.42% respectively. The Combined dataset, which integrated the strengths of both datasets, exhibited a stronger generalization with an accuracy of 97.48% and validation accuracy of 97.49%, balancing the need for high-quality images with environmental variability. The study shows improved CLR detection speed and accuracy due to high-quality and consistently labeled images from the Mbozi dataset. It recommends that future models integrate regionally relevant and high resolution datasets for robust performance in real-world agricultural conditions, providing coffee farmers with timely disease intervention tools for better production management and economic stability in coffee-growing regions.
  • Thumbnail Image
    Item
    Group-Based Data Offloading Techniques Assisted by D2D Communication in 5G Mobile Network
    (Tanzania Journal of Engineering and Technology, 2022-07-30) Ally, Juma Said
    Machine type communication devices proposed as one of the substantial data collections in the 5G of wireless networks. However, the existing mobile communication network is not designed to handle massive access from the MTC devices instead of human type communication. In this context, we propose the device-to-device communication assisted a mobile terminal (smartphone) on data computing, focusing on data generated from a correlated source of machine type communication devices. We consider the scenario that the MTC devices after collecting the data will transmit to a smartphone for computing. With the limitation of computing resources at the smartphone, some data are offloaded to the nearby mobile edge-computing server. By adopting the sensing capability on MTC devices, we use a power exponential function to compute a correlation coefficient existing between the devices. Then we propose two grouping techniques K-Means and hierarchical clustering to combine only the MTC devices, which are spatially correlated. Based on this framework, we compare the energy consumption when all data processed locally at a smartphone or remotely at mobile edge computing server with optimal solution obtained by exhaustive search method. The results illustrated that; the proposed grouping technique reduce the energy consumption at a smartphone while satisfying a required completion time.