Integrating a tethered Remotely Operated Vehicle (ROV) with the YOLOv7 object recognition algorithm offers a novel approach to marine monitoring, particularly for tracking giant clam populations in the reefs of Tioman Island, Malaysia. The ROV, designed for enhanced underwater exploration, is equipped with advanced imaging and sonar technologies, resilient tethered power management and communication systems. The control and navigation system are built on the pixhawk flight controller, supported by ArduPilot firmware and a customized QGroundControl interface, ensuring real-time operations. The primary objective of this research is to develop an algorithm capable of accurately recognizing and counting giant clams through underwater images and videos. The study compares the performance of the YOLOv7 algorithm with the manual counting. The findings indicate that the YOLOv7 algorithm demonstrates average accuracy of the model of 79.65%. The integration of YOLOv7 enables identification and quantification of giant clam populations enhancing data collection for research and conservation efforts providing practical solutions for environmental monitoring and the sustainable management of marine resources.