Urban Land Cover Classification with Missing Data Modalities Using Deep Convolutional Neural Networks

TL;DR

This paper introduces a CNN architecture with hallucination networks that effectively handles missing data modalities in urban land cover classification, improving performance when some sensor data is unavailable at test time.

Contribution

The proposed CNN with hallucination networks enables fusion of multiple data modalities even when some are missing during testing, a capability lacking in current methods.

Findings

Outperforms standard CNNs trained on optical images.

Better than ensemble of separate CNNs.

Effective with partial missing modalities during testing.

Abstract

Automatic urban land cover classification is a fundamental problem in remote sensing, e.g. for environmental monitoring. The problem is highly challenging, as classes generally have high inter-class and low intra-class variance. Techniques to improve urban land cover classification performance in remote sensing include fusion of data from different sensors with different data modalities. However, such techniques require all modalities to be available to the classifier in the decision-making process, i.e. at test time, as well as in training. If a data modality is missing at test time, current state-of-the-art approaches have in general no procedure available for exploiting information from these modalities. This represents a waste of potentially useful information. We propose as a remedy a convolutional neural network (CNN) architecture for urban land cover classification which is able to embed all available training modalities in a so-called hallucination network. The network will in effect replace missing data modalities in the test phase, enabling fusion capabilities even when data modalities are missing in testing. We demonstrate the method using two datasets consisting of optical and digital surface model (DSM) images. We simulate missing modalities by assuming that DSM images are missing during testing. Our method outperforms both standard CNNs trained only on optical images as well as an ensemble of two standard CNNs. We further evaluate the potential of our method to handle situations where only some DSM images are missing during testing. Overall, we show that we can clearly exploit training time information of the missing modality during testing.