Date of Publication

3-2025

Document Type

Bachelor's Thesis

Degree Name

Bachelor of Science in Manufacturing Eng'g & Mgt w/ Specialization in Mechatronics & Robotics Eng'g

Subject Categories

Biomedical Engineering and Bioengineering

College

Gokongwei College of Engineering

Department/Unit

Manufacturing Engineering and Management

Thesis Advisor

Ira C. Valenzuela

Defense Panel Chair

Ronnie S. Concepcion II

Defense Panel Member

Robert Kerwin C. Billones
Renann G. Baldovino

Abstract (English)

Abstract— Reputed as the rainforests of the sea, coral reefs were an essential part of the marine ecosystem; they were known to be home to most benthic life forms, which were an integral part of different anthropogenic activities, including commercial fishing and tourism. However, marine reefs were under constant siege due to environmental pressures such as the growth of Crown-of-Thorns Starfish (CoTS). This research focused on developing a camera device prototype with an algorithm that identified the population of the Crown-of-Thorns Starfish within a 1 x 1 m² sampling station in a controlled environment. This organism was relatively invasive as it latched onto the surface of the reef and eventually encased corals, decimating 90% of them through the secretion of an enzyme that specialized in breaking down polyps. The Crown-of-Thorns Starfish were also reported to be insatiable predators that could reach maturity in less than a year and reproduce proficiently, posing a definite threat to corals, especially in their matured form. Therefore, this emphasized the importance of regulating the Crown-of-Thorns Starfish to preserve coral reefs. The Crown-of-Thorns Starfish was an ongoing threat to reef-building corals that inhibited the growth of the fish population. The study focused on mapping the Crown-of-Thorns Starfish’s specific coordinates with the use of Neo-6m - a GPS module. Apart from that, the study created a model to predict the radius which was determined by the pixel difference of bounding boxes and estimate population density through unique ID counting to determine the necessary course of action for rehabilitating the marine ecosystem. This approach integrated the highest-performing models—Custom CNN, VGG-19, ResNet - 50, MobileNet, and YOLOv8n—as the algorithm for object detection and DeepSORT for localization. The data was also reflected and visualized through Thingsboard IoT for real-time and remote monitoring. The proposed systems held great potential in conservation efforts due to timely intervention procedures in mitigating the impact of the Crown-of-Thorns Starfish on the designated ecosystem.

Abstract Format

html

Abstract (Filipino)

None

Abstract Format

html

Language

English

Format

Electronic

Keywords

Crown-of-thorns starfish; Environmental monitoring

Upload Full Text

wf_yes

Embargo Period

3-2026

Download

Share

COinS