fish-detection

Welcome to the Fish-Detection-YOLOv8 repository! This project focuses on detecting fish species using the YOLOv8 object detection algorithm. It aims to provide accurate and efficient fish detection models for applications such as marine biology research and automated fishing systems.

A robust computer vision pipeline for underwater fish analysis. It uses a YOLOv8 model for stable, real-time tracking and classification. Features multiple processing modes, including buffered real-time analysis and high-accuracy offline filtering, making it a flexible tool for marine biologists and researchers.

This code is a PyTorch implementation of ClassAwareLoss proposed in the "Class-aware fish species recognition using deep learning for an imbalanced dataset" paper. https://www.mdpi.com/1424-8220/22/21/8268

Deep learning fish classifier combining ConvNeXt-Tiny (40 species, 98.96% accuracy) with BioCLIP-2 zero-shot recognition and AI-powered habitat mapping

This project is a collaborative fog node-based fish detection system. It leverages fog computing to distribute image processing tasks across multiple clients and a server. The system is designed to detect fish in images and provide bounding box information using the Roboflow API.

Utilizes Deep Learning and Image Processing techniques to perform image restoration and enhancement and further apply fish / coral detection.

This project is a collaborative fog node-based fish detection system. It leverages fog computing to distribute image processing tasks across multiple clients and a server. The system is designed to detect fish in images and provide bounding box information using the Roboflow API.

Use local AI for object detection to sort and crop photos.

ORCA: Oceanic Recognition & Classification Application for sea-life analysis systems.

FishTrack is an autonomous, solar-powered marine monitoring buoy system that integrates sonar, edge-AI fish identification, GPS tracking, and a custom LoRa multi-hop mesh network to dynamically map marine environments and broadcast coordinates to fishers without requiring internet connectivity.

ML app for early prediction Fish disease using Water Quality instead of images

AI aquarium vision monitoring MVP using Raspberry Pi, fish detection, trajectory tracking, and behavior-based abnormal activity alerts.

This model is a convolutional neural network trained to remove underwater "clicks" from crustacea, focused on cleaning signals from fish.

This project details the process of training a custom YOLOv5 object detection model on a dataset of salmonid fish. The final model is converted to the TensorFlow Lite (TFLite) format, making it lightweight and optimized for real-time inference in mobile applications.

Multi-Hop LoRa Mesh Network & AI-Powered Oceanographic Monitoring Control Center. Integrates edge YOLO fish detection with BFAR MIMAROPA taxonomical database and Gemini AI verification, sonar telemetry reassembly, and a real-time Flask command dashboard.