Satellite Image Segmenation

Problem statement and Objective

Aim of this project is to propose and implement a satellite image classification technique for a given data set of satellite imagery into 9 categories namely, Roads, Buildings, Trees, Grass, Bare Soil, Water, Railways, Swimming pools and Unlabelled areas.

Table of contents

• Satellite Image Segmenation
  • Problem statement and Objective
• Table of contents
• Approach
• Existing work
  • Standard U-net Model
  • Vegetation and Aqua Indices
  • Keras(TensorFlow backend)
  • Optimizer and Activation functions
    • Cost Function
      • Evaluation metrics
  • Experiments
    • Larger U-net with 8-channel input
    • Double U-net with a larger standard and smaller incorporating indices
    • Threshold variations and learning rate variation
    • Statistics
    • Processing
• Implementation
  • Model and Architecture
  • Pipeline
    • Final prediction
    • Testing
• Novelty
  • Unique Network Architecture Design
  • Reflection Padding for Global Boundary loss minimisation
  • Overlapping Patches and Selection Procedure for Local Boundary loss minimisation
  • Rotate Augmentation for Smooth Segmentation
• Results
  • Accuracy and Cohen Kappa Coefficient
  • Confusion Matrix
• Conclusion
• Credits

Approach

When we visualized the problem statement and the dataset, we tried to find resources related to the same and also started analysing the techniques which could be applied for the solution. On the first go we foundthat the dataset has some missing labels meaning where there is clearly visible that a particular class has to be present but in ground truth it is unlabelled.

Existing work

Work has been done before in this domain of segmenting satellite imagery, using a number of techniques. Various versions of Convolutional Neural Networks are observed to be used to solve this and related problems. The use of the special type of CNN termed as U-net for image segmentation was first observed in the paper titled U-Net: Convolutional Networks for Biomedical Image Segmentation. In the next 3 years, this technique has also found its applications in related domains like satellite imagery.

Standard U-net Model

Vegetation and Aqua Indices

Keras(TensorFlow backend)

Keras

Optimizer and Activation functions

Cost Function

Evaluation metrics

Experiments

Larger U-net with 8-channel input

Double U-net with a larger standard and smaller incorporating indices

Threshold variations and learning rate variation

Statistics

Processing

Implementation

Model and Architecture

Pipeline

Final prediction

Testing

Novelty

Unique Network Architecture Design

Reflection Padding for Global Boundary loss minimisation

Overlapping Patches and Selection Procedure for Local Boundary loss minimisation

Rotate Augmentation for Smooth Segmentation

Results

Accuracy and Cohen Kappa Coefficient

Confusion Matrix

Conclusion

Credits

• Piyush Chauhan
• Milan Jolly
• Parth Kulkarni

---