The project focuses on developing a fish detection system using advanced object detection models based on the YOLO architecture. By utilizing deep learning models, specifically YOLOv11 and YOLOv9, the system is designed to detect fish species in underwater images. The models are trained on a diverse dataset containing various fish species found in marine environments. The primary goal of the system is to offer an efficient, accurate, and scalable solution for fish detection in both still images and live video streams. A web application built with Python Flask serves as the platform for users to interact with the detection system, allowing them to upload images or stream live video for species classification. The application provides easy-to-use interfaces for registration, login, image upload, and live detection. The project aims to contribute to marine research by providing a tool that supports accurate monitoring of aquatic life, aiding in the study and preservation of marine ecosystems.

Keywords: Fish detection, YOLOv11, YOLOv9, object detection, marine ecosystems, deep learning, Python Flask, image classification, underwater images, live video detection.