Analysis of Ice-Ice Disease on Seaweed Using the K-Nearest Neighbor Algorithm in Vision Robot Technology

Abstract

The advancement of remote sensing technology has significantly enhanced marine monitoring, particularly in assessing the quality of seaweed—a vital natural resource for various industries. Traditional seaweed monitoring methods often lack precision in disease detection. This study introduces a refined approach using the K-Nearest Neighbors (KNN) algorithm, tailored for remote imagery and sensor-based seaweed disease identification. Two key aspects are emphasized: first, the classification of seaweed diseases, critical for maintaining growth and quality; and second, detection based on color and size analysis to ensure productivity. By utilizing image analysis of detection vessels, the system effectively identifies diseases, particularly Ice-Ice disease, optimizing treatment strategies for seaweed farmers. The research demonstrated the system's capability through 44 tests, achieving an accuracy rate of 86.67%. This innovation significantly boosts production efficiency and seaweed quality, providing substantial benefits for sustainable seaweed farming practices. This study highlights the potential of integrating IoT and machine learning technologies to support precise, efficient disease detection and monitoring, contributing to the sustainability and economic viability of the seaweed industry.