Convolutional variational autoencoders for secure lossy image compression in remote sensing

TL;DR

This work tackles onboard secure lossy compression of large-scale remote-sensing imagery using convolutional variational autoencoders with an entropy bottleneck to optimize a rate-distortion objective. The approach learns compact latent representations that can be securely transmitted, with the encoder deployed on-board and the decoder on the ground. Experiments on the GRSS dataset show the vanilla CVAE achieves substantial compression while preserving reconstruction quality (MSE, SSIM, PSNR), delivering favorable rate-distortion trade-offs relative to traditional methods. The findings support practical deployment on small satellites and edge platforms, offering a lightweight, secure compression pathway for high-volume remote-sensing data.

Abstract

The volume of remote sensing data is experiencing rapid growth, primarily due to the plethora of space and air platforms equipped with an array of sensors. Due to limited hardware and battery constraints the data is transmitted back to Earth for processing. The large amounts of data along with security concerns call for new compression and encryption techniques capable of preserving reconstruction quality while minimizing the transmission cost of this data back to Earth. This study investigates image compression based on convolutional variational autoencoders (CVAE), which are capable of substantially reducing the volume of transmitted data while guaranteeing secure lossy image reconstruction. CVAEs have been demonstrated to outperform conventional compression methods such as JPEG2000 by a substantial margin on compression benchmark datasets. The proposed model draws on the strength of the CVAEs capability to abstract data into highly insightful latent spaces, and combining it with the utilization of an entropy bottleneck is capable of finding an optimal balance between compressibility and reconstruction quality. The balance is reached by optimizing over a composite loss function that represents the rate-distortion curve.

Figures (7)

• Figure 1: Schematic diagram of HSOCVQ method.qian_optical_2013
• Figure 2: Schematic diagram of image compression.
• Figure 3: Variational Autoencoder architecture schematic murphy_probabilistic_2022.
• Figure 4: Samples from the GRSS dataset.
• Figure 5: VAE architecture used.