Lightweightability Measurement and a General Lightweight Design Framework for On-Orbit Image Interpretation Neural Networks

Satellite on-orbit remote sensing image intelligent interpretation relies on on-orbit devices with extremely limited computational resources, using advanced neural networks designed through lightweight methodologies to achieve fast and accurate interpretation of on-orbit remote sensing images. Nevertheless, the current design of lightweight neural networks exhibits three salient issues: first, the prevailing “one-size-fits-all” network lightweight design pattern is notably inefficient; second, there is a deficiency in analyzing the impact of lightweight operations on network performance; third, the mutual influence among various lightweight operations is overlooked. To address these issues, first, we propose a neural network lightweightability measurement model and its computational method by investigating the effects of various lightweight operations; second, we propose a neural network general lightweight design (GLD) framework tailored for satellite on-orbit remote sensing images intelligent interpretation. Specifically, GLD, based on a meta-leaning approach, integrates knowledge distillation (KD), pruning, and quantization, three general lightweight technologies, into a framework. It dynamically assesses the distillability, prunability, and quantifiability of neural networks and uses this assessment and uses this as supervision to dynamically jointly optimizes KD, pruning, and quantization (KDPQ), making it applicable to various mainstream neural networks; furthermore, we explore the mutual influence among lightweight operations based on GLD; finally, through ablation experiments and comparative experiments, we further verify the effectiveness and superiority of GLD.