| dc.description.abstract | Coconut production is a cornerstone of agriculture in tropical regions, significantly
contributing to national economies. However, coconut trees are increasingly vulnerable
to pests and diseases, which critically impact productivity. This research proposes a
comprehensive framework for the early detection of major coconut diseases, including
stem bleeding, bud rot, and bud root dropping, as well as pest infestations, particularly
from coconut caterpillars. Utilizing image processing and deep learning technologies,
the study focuses on the stem and bud regions, achieving a detection accuracy exceeding
90%. The research methodology incorporates a systematic literature review and surveys
to analyze existing detection techniques and management practices. The literature
review identifies gaps in current studies, revealing that fewer than 20% focus specifically
on stem and bud regions and that over 75% rely on traditional methods. Surveys with
farmers and agricultural experts provide insights into practical challenges in disease
detection and pest management, guiding the design of the proposed framework. This
integrated approach not only ensures timely and accurate identification but also reduces
disease and pest spread by up to 50% compared to traditional methods. The study also
emphasizes pest prevention strategies and practical management solutions, combining
biological and chemical control measures. The research highlights the critical need
for advanced technologies such as deep learning and real-time monitoring for early
detection and integrated management. It provides actionable management strategies
to mitigate damage, enhance plantation health, and improve overall productivity.
This work serves as a valuable resource for farmers, researchers, and policymakers,
contributing to the protection and sustainability of coconut cultivation against pests
and diseases. | en_US |
