Reducing the feature divergence of RGB and near-infrared images using Switchable Normalization

TL;DR

This paper introduces a method using Switchable Normalization in a segmentation model to reduce feature divergence between RGB and near-infrared images, leading to significant improvements in agricultural aerial image analysis.

Contribution

The paper proposes applying Switchable Normalization to address feature divergence in multi-modality aerial images, improving segmentation performance.

Findings

Reduces feature divergence between RGB and NIR images.

Achieves nearly 10% improvement in mean IoU.

Effectively integrates multi-modality data for better segmentation.

Abstract

Visual pattern recognition over agricultural areas is an important application of aerial image processing. In this paper, we consider the multi-modality nature of agricultural aerial images and show that naively combining different modalities together without taking the feature divergence into account can lead to sub-optimal results. Thus, we apply a Switchable Normalization block to our DeepLabV3 segmentation model to alleviate the feature divergence. Using the popular symmetric Kullback Leibler divergence measure, we show that our model can greatly reduce the divergence between RGB and near-infrared channels. Together with a hybrid loss function, our model achieves nearly 10\% improvements in mean IoU over previously published baseline.