# Unsupervised Change Detection in Satellite Images Using Convolutional Neural Networks

## TL;DR

This paper introduces an unsupervised change detection method for satellite images that leverages pre-trained CNNs for feature extraction and semantic classification, eliminating the need for labeled difference images.

## Contribution

It presents a novel unsupervised approach that uses CNN feature maps to detect and classify changes without explicit training on difference images.

## Key findings

- Effective change detection without labeled difference data
- Compatible with any pre-trained CNN for semantic segmentation
- Achieves accurate change localization and classification

## Abstract

This paper proposes an efficient unsupervised method for detecting relevant changes between two temporally different images of the same scene. A convolutional neural network (CNN) for semantic segmentation is implemented to extract compressed image features, as well as to classify the detected changes into the correct semantic classes. A difference image is created using the feature map information generated by the CNN, without explicitly training on target difference images. Thus, the proposed change detection method is unsupervised, and can be performed using any CNN model pre-trained for semantic segmentation.