Yolo11 Fish Detector Grayscale

🚀 Yolo11n Fish Detector | Grayscale

This model is designed to detect fish in underwater grayscale imagery. It leverages the YOLO11 architecture and semi - unsupervised learning techniques, enabling it to identify fish without relying on fully labeled datasets.

✨ Features

• Model Architecture: YOLO11n, a lightweight architecture optimized for real - time fish detection in underwater footage.
• Task: Object Detection, specifically focused on fish detection.
• Footage Type: It is trained on grayscale (black - and - white) underwater footage.
• Classes: The model can detect 1 class, which is fish.

📦 Installation

No specific installation steps are provided in the original document, so this section is skipped.

💻 Usage Examples

Basic Usage

To use the trained model, follow these steps:

from ultralytics import YOLO

# Load the model
model = YOLO("yolo11n_fish_trained.pt")

📚 Documentation

Model Details / Overview

This model was trained to detect fish in underwater Grayscale Imagery using the YOLO11 architecture. The model was developed using semi - unsupervised learning techniques to learn patterns and identify fish without relying on fully labeled datasets.

Test Results

Model Weights

The model's weights can be found here

Intended Use

• Real - time detections on Grayscale (Black - and - White) Underwater Footage
• Post - processed video/imagery on Grayscale (Black - and - White) Underwater Footage

Factors

Model Performance

• Unsupervised Learning: It allows for flexible pattern recognition without fully annotated datasets, which is well - suited for grayscale imagery.
• Grayscale Imagery: The model is trained specifically on black - and - white underwater footage, enhancing its performance in this environment.
• Model Architecture (YOLO11n): It is lightweight and optimized for real - time fish detection in underwater footage.
• Training Data: The quality and diversity of the grayscale underwater dataset, split into 80% training and 20% validation, contribute to the model's performance.
• Training Parameters: Configured with 50 epochs, a 0.001 learning rate, and 416x416 image size for optimal model convergence.

Metrics

Below are the key metrics from the model evaluation on the validation set:

Training Validation Results

Training and Validation Losses

Confusion Matrix

Precision - Recall Curve

F1 Score Curve

Training Configuration

• Model Weights File: yolo11n_fish_trained.pt
• Number of Epochs: 50
• Learning Rate: 0.001
• Batch Size: 16
• Image Size: 416x416

Training and Evaluation Data

• Dataset: fish_dataset.zip consisting of Grayscale (black - and - white) underwater footage.
• Training/Validation Split: The dataset was split into 80% training and 20% validation.
• Classes: The model was trained on a single class (fish).
• Learning Approach: Unsupervised learning, meaning the model identified patterns in the data without needing detailed annotations for all images.

Deployment

How to Use the Model

To use the trained model, follow these steps:

from ultralytics import YOLO

# Load the model
model = YOLO("yolo11n_fish_trained.pt")

Limitations

• The model was trained on black - and - white underwater footage, and may not generalize well to color images or videos with different lighting conditions.
• The unsupervised learning nature of this model may lead to some incorrect detections, particularly in noisy environments where it may confuse other underwater objects for fish.
• Images with complex backgrounds, occlusions, or poor resolution may affect the model's performance.

Additional Notes

• Grayscale Imagery: The model may perform better on grayscale images and might not generalize well to color underwater footage or images with different lighting conditions.
• Unsupervised Learning: Since it uses an unsupervised approach, it's worth noting that this can make the model more flexible but also more prone to errors or misclassifications without annotated data.
• Ethical Considerations: The unsupervised learning approach could lead to biases in detections, especially in new environments or types of marine life that were not represented in the training dataset. This model should not be used in critical applications without thorough validation to ensure it doesn't miss key detections or produce incorrect results in sensitive scenarios. Consider the potential environmental or societal impact when using the model for marine conservation or research, and ensure that the detections are verified.

Disclaimer

This repository is a scientific product and is not official communication of the National Oceanic and Atmospheric Administration, or the United States Department of Commerce. All NOAA project content is provided on an ‘as is’ basis and the user assumes responsibility for its use. Any claims against the Department of Commerce or Department of Commerce bureaus stemming from the use of this project will be governed by all applicable Federal law. Any reference to specific commercial products, processes, or services by service mark, trademark, manufacturer, or otherwise, does not constitute or imply their endorsement, recommendation or favoring by the Department of Commerce. The Department of Commerce seal and logo, or the seal and logo of a DOC bureau, shall not be used in any manner to imply endorsement of any commercial product or activity by DOC or the United States Government.

📄 License

This model is licensed under the AGPL - 3.0 license.