XAI based Performance Preserving Adaptive Image Compression for Efficient Satellite Communication

TL;DR

This paper introduces RDIC, an adaptive image compression method that preserves important satellite image regions based on analysis model insights, reducing transmission overhead without sacrificing analysis accuracy.

Contribution

The paper presents a novel reasoning-based image compression scheme that dynamically compresses images by leveraging pixel importance scores from analysis models.

Findings

High compression rate achieved with low accuracy loss

Effectively captures important image regions

Reduces satellite data transmission overhead

Abstract

In the era of multinational cooperation, gathering and analyzing the satellite images are getting easier and more important. Typical procedure of the satellite image analysis include transmission of the bulky image data from satellite to the ground producing significant overhead. To reduce the amount of the transmission overhead while making no harm to the analysis result, we propose a novel image compression scheme RDIC in this paper. RDIC is a reasoning based image compression scheme that compresses an image according to the pixel importance score acquired from the analysis model itself. From the experimental results we showed that our RDIC scheme successfully captures the important regions in an image showing high compression rate and low accuracy loss.