# Canonical Correlation Analysis for Misaligned Satellite Image Change Detection

## TL;DR

This paper introduces an alignment-agnostic CCA method that effectively detects changes in multi-temporal satellite images despite misalignment errors, improving robustness in multi-view data analysis.

## Contribution

The paper proposes a novel AA CCA variant that mitigates alignment errors, enhancing change detection accuracy in satellite imagery with mispaired data.

## Key findings

- AA CCA outperforms traditional CCA in misaligned data scenarios
- The method improves change detection accuracy in satellite images
- AA CCA demonstrates robustness to alignment errors

## Abstract

Canonical correlation analysis (CCA) is a statistical learning method that seeks to build view-independent latent representations from multi-view data. This method has been successfully applied to several pattern analysis tasks such as image-to-text mapping and view-invariant object/action recognition. However, this success is highly dependent on the quality of data pairing (i.e., alignments) and mispairing adversely affects the generalization ability of the learned CCA representations. In this paper, we address the issue of alignment errors using a new variant of canonical correlation analysis referred to as alignment-agnostic (AA) CCA. Starting from erroneously paired data taken from different views, this CCA finds transformation matrices by optimizing a constrained maximization problem that mixes a data correlation term with context regularization; the particular design of these two terms mitigates the effect of alignment errors when learning the CCA transformations. Experiments conducted on multi-view tasks, including multi-temporal satellite image change detection, show that our AA CCA method is highly effective and resilient to mispairing errors.