← Back to the feed
Space & aviationResearch findings

SCI-Mamba: A Novel Unsupervised Low-Light Image Enhancement Method for Spacecraft

Original published 2026.07.09 10:17 KSTAdded to Blackdeck 2026.11.20 09:20 KST

What has changed?

A lightweight, unsupervised framework combining self-calibrated learning and Retinex physical priors for low-light enhancement

Key points

  1. Proposes SCI-Mamba, an unsupervised low-light enhancement method for spacecraft imaging.
  2. Validated using Space Dark-1.0 dataset covering diverse illumination and motion conditions.
  3. Achieves superior visual authenticity and inference speed compared to CNN and Mamba-based methods.

Connect it to your work

The method enables real-time processing suitable for resource-constrained spaceborne systems

What remains unproven

This research note is based on an abstract; check the original for test conditions and application scope.

Originals used for this note

Links open the original publisher in a new tab.

  1. arxivSCI-Mamba: Unsupervised Learning based Low-Light Image Enhancement for Non-Cooperative Spacecraft