Journal of Computer Networks, Architecture and High Performance Computing
Vol. 5 No. 1 (2023): Article Research Volume 5 Issue 1, January 2023

Implementation of ColorSpace, GrabCut, and Watershed Methods on Digital Image Segmentation of Coral and Fish Objects

Aditya Ahmad Fauzi (STIE PERTIBA)
Adisuputra (STIE PERTIBA)



Article Info

Publish Date
24 Jan 2023

Abstract

Poor coral reefs rising in eastern Indonesia. The colors have disappeared from the seafloor instead, bleached branches are visible. To recognize the differences between dead and healthy coral reefs, an identification system has been created using the image processing method. Object segmentation is a step in digital image processing to separate one object from another based on specific characteristics. In this study, coral objects with various colored backgrounds and things became a problem to separate, so this study aimed to separate these various colors. This research uses color space segmentation to visualize RGB and HSV colors, Grabcut segmentation to separate the largest corals, and watershed segmentation to separate dead corals. Therefore, from this study, the RGB and HSV color visualizations were clearly visible. From Grabcut segmentation, it is found that the largest fish is detected and can be displayed in the segmentation results. At the same time, the watershed segmentation displays dead coral taken by gray segmentation with otsu.

Copyrights © 2023






Journal Info

Abbrev

CNAPC

Publisher

Subject

Computer Science & IT Education

Description

Journal of Computer Networks, Architecture and Performance Computing is a scientific journal that contains all the results of research by lecturers, researchers, especially in the fields of computer networks, computer architecture, computing. this journal is published by Information Technology and ...