<![CDATA[Advances in digital image processing technology and machine learning, such as clustering, have contributed to increased efficiency in various sectors, including marine and fisheries. Octopus, lobsters, and shellfish are high-value fishery commodities that have traditionally been classified manually, with the potential for subjectivity and inefficiency. This study aims to develop a digital image classification model for marine biota using the K-Means Clustering method equipped with image processing techniques. The methods applied include converting the RGB color space to L*a*b, segmentation with K-means, shape feature extraction (metric, eccentricity) and GLCM texture (contrast, correlation, energy, homogeneity). The results show that this method is effective in identifying the three types of marine biota with an average accuracy of 95% based on testing on 30 images. The implementation of K-means Clustering has been proven to be accurate and consistent in the automation of marine biota classification.]]>
Copyrights © 2025
