Published Paper Details:

Underwater object detection is a critical aspect of various marine applications including environmental monitoring, underwater exploration and infrastructure maintenance. However traditional methods encounter challenges such as poor visibility complex backgrounds and limited computational resources. Deep learning techniques particularly the YOLO model have shown promise in addressing these challenges by enabling rapid and accurate object detection in real-time. In this study we propose a novel approach for underwater object detection using the YOLO model augmented with an alert mechanism to notify users of detected objects. The proposed system leverages the YOLO model's ability to detect objects with high precision and efficiency. By training the model on underwater image datasets, it learns to recognize various objects commonly found in underwater environments including marine life structures and debris. Additionally we integrate a buzzer alert system into the detection pipeline enhancing the system's usability and practicality for real-world applications. During operation, the system continuously analyzes live underwater video feeds captured by underwater cameras. Upon detecting an object of interest such as marine debris or an unfamiliar structure the YOLO model triggers the buzzer alert mechanism. This alerts users such as marine researchers environmentalists or underwater vehicle operators, to the presence of the detected object in real-time. Users can then take appropriate actions such as investigating the object further or adjusting the trajectory of underwater vehicles to avoid collisions.