Spectral Image Visualization Using Generative Adversarial Networks

TL;DR

This paper introduces a GAN-based method for visualizing spectral images in natural colors, preserving structure and conveying more information than traditional false-color methods.

Contribution

The paper proposes a novel GAN framework with a combined loss function for natural color visualization of spectral images, addressing limitations of existing false-color techniques.

Findings

Generates natural-looking spectral image visualizations

Preserves structural information in the images

Outperforms traditional false-color visualization methods

Abstract

Spectral images captured by satellites and radio-telescopes are analyzed to obtain information about geological compositions distributions, distant asters as well as undersea terrain. Spectral images usually contain tens to hundreds of continuous narrow spectral bands and are widely used in various fields. But the vast majority of those image signals are beyond the visible range, which calls for special visualization technique. The visualizations of spectral images shall convey as much information as possible from the original signal and facilitate image interpretation. However, most of the existing visualizatio methods display spectral images in false colors, which contradict with human's experience and expectation. In this paper, we present a novel visualization generative adversarial network (GAN) to display spectral images in natural colors. To achieve our goal, we propose a loss function which consists of an adversarial loss and a structure loss. The adversarial loss pushes our solution to the natural image distribution using a discriminator network that is trained to differentiate between false-color images and natural-color images. We also use a cycle loss as the structure constraint to guarantee structure consistency. Experimental results show that our method is able to generate structure-preserved and natural-looking visualizations.