oil-spill-tiyas

DeepLabV3+ Oil Spill Detection - Oil Spill Detection

Model Description

High-accuracy DeepLabV3+ model for oil spill detection from satellite images. Best performance for critical applications.

Model Details

• Model Size: 204.56 MB
• F1 Score: 0.9668
• Input Shape: (256, 256, 3) - RGB satellite images
• Output Shape: (256, 256, 5) - 5-class semantic segmentation
• Classes: Background, Oil Spill, Ships, Look-alike, Wakes
• Format: TensorFlow Keras (.keras)

Usage

import tensorflow as tf
import numpy as np
from PIL import Image

# Load the model
model = tf.keras.models.load_model('model.keras', compile=False)

# Preprocess image
def preprocess_image(image_path):
    image = Image.open(image_path).convert('RGB')
    image = image.resize((256, 256))
    img_array = np.array(image) / 255.0
    img_array = np.expand_dims(img_array, axis=0)
    return img_array

# Make prediction
image_array = preprocess_image('your_image.jpg')
prediction = model.predict(image_array)

Model Performance

This model achieves an F1 score of 0.9668 on the oil spill detection task, making it suitable for production deployment in environmental monitoring systems.

Deployment

This model is optimized for production deployment with:

• Fast inference times
• Memory-efficient architecture
• Support for batch processing
• Compatible with TensorFlow Serving

License

MIT License - Free for academic and commercial use.