SatReg: Regression-based Neural Architecture Search for Lightweight Satellite Image Segmentation

2026-04-11

AI Summary

This work addresses the challenge of deploying lightweight remote sensing image segmentation models under stringent latency and energy constraints in on-orbit and edge computing scenarios. We propose a hardware-aware neural architecture search method that compresses the high-dimensional search space into two critical width variables. Leveraging CM-UNet as a teacher model, our approach integrates knowledge distillation with a low-fidelity regression surrogate model sampled on the Jetson Orin Nano to jointly predict segmentation accuracy and hardware costs, enabling rapid architecture selection. The method substantially improves search efficiency, achieving competitive mean Intersection-over-Union (mIoU) while significantly reducing latency and power consumption. Experimental results validate the effectiveness of the reduced-space regression strategy tailored for CNN-Mamba hybrid architectures in satellite-edge deployment settings.

Technology Category

• Machine Learning: Hardware-aware ML
• Computer Vision: Segmentation
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

As Earth-observation workloads move toward onboard and edge processing, remote-sensing segmentation models must operate under tight latency and energy constraints. We present SatReg, a regression-based hardware-aware tuning framework for lightweight remote-sensing segmentation on edge platforms. Using CM-UNet as the teacher architecture, we reduce the search space to two dominant width-related variables, profile a small set of student models on an NVIDIA Jetson Orin Nano, and fit low-order surrogate models for mIoU, latency, and power. Knowledge distillation is used to efficiently train the sampled students. The learned surrogates enable fast selection of near-optimal architecture settings for deployment targets without exhaustive search. Results show that the selected variables affect task accuracy and hardware cost differently, making reduced-space regression a practical strategy for adapting hybrid CNN-Mamba segmentation models to future space-edge systems.

Problem

• satellite image segmentation
• neural architecture search
• edge computing
• hardware-aware
• lightweight models

Innovation

• neural architecture search
• hardware-aware optimization
• knowledge distillation
• surrogate modeling
• edge computing