Convolutional Neural Network (CNN) is a set of deep learning algorithms that have been widely used in many scenarios and different practices, especially in the area of image processing and analysis, including image classification, object detection, and segmentation. In this research project, the authors tried to employ a popular application of the CNN, the YOLO algorithm, to perform object detection for the video images obtained at the beach coast of the Pacific Island, Guam. The purpose of this initiative is to establish an automatic environmental monitoring system for beach coastal safety and life rescue for the tourists who visit Guam and swim in the ocean waters of the beaches of Guam. The CNN-based AI algorithm, YOLO, is highly effective for real-time video image processing due to YOLO’s fast speed and efficiency in image analysis and object detection. In this paper, the authors will describe how they selected the ocean surface conditions and took the video images, as well as how they applied YOLO algorithms to obtain the object detection parameter, the numeric values of the confidence threshold. Based on the analysis of the YOLO generated confidence threshold values, a number of interesting findings have been made for the coastal safety monitoring on Guam’s beaches.

CNN-Based YOLO, Beach Coast Safety, Video Image Analysis, Object Detection, Guam