Skip to content

japhethkimeu/deforestation_detection_using_deep_learning

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

15 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Detecting deforestation using deep learning

This projet entails the application of deep learning algorithms i.e. CNN to detect deforestation using satellite imagery.

Data

I analysed Sentinel-2 images (cloud free) of three different regions in the State of Mato Grosso in Brazil and derived a total of 5,122 image chips and masks of size 256 * 256. See a sample of training images and masks in the sample data folder.

Data Augmentation

Image chips were flipped left and right before fed into the network for training. See preprocess.py

Model

Unet model was implemented by using MobileNetV2 as the feature extraction with pre-trained weights.

results/u-net-architecture.png

More on Unet model U-Net: Convolutional Networks for Biomedical Image Segmentation

Training

The model was trained first using the pre-trained weights of the base model for 15 epochs with early stopping. The best result was achieved by re-training the whole model for 20 epochs and attained a binary accuracy of 0.9780.

Requirements

    - tensorflow
    - keras
    - numpy
    - matplotlib
    - pandas
    - cv2

Results

results/predict_2.PNG