稀疏地刻画变化：面向遥感解译的脉冲时序框架
Sparsely Timing the Change: A Spiking Temporal Framework for Remote Sensing Interpretation
把双时相辐射差转成带 time-to-first-spike 特性的稀疏脉冲序列，再用 SNN 提取动态变化特征，并通过可学习时序衰减与 ANN 语义特征融合——这套「脉冲编码+轻量适配器」思路可迁移到任何需要从稀疏时序观测中挖掘动态信息的视觉任务（如视频变化检测、事件相机处理）。
原文依据 transforms bi-temporal radiometric differences into sparse spike sequences with time-to-first-spike characteristics
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论地表空间结构的时序演化模式是智能遥感解译领域的核心关注点。然而，受限于仅能获取两个时相的数据，如何对稀疏的时空变化过程建模以有效解译地表变化，仍是智能遥感分析中的核心挑战。为此，本文提出 SpikeAdapter，一个轻量级增强框架。该框架由 Geo-Spike 插值模块（GSI-P）、脉冲神经网络（SNN）特征提取器以及时空融合模块 STSpikeFuse 组成。受大脑对新异与消退刺激的感知响应启发，核心的 GSI-P 模块将双时相辐射差转换为具有首次脉冲时间（time-to-first-spike）特性的稀疏脉冲序列。随后，我们使用 SNN 特征提取器捕获地表目标的动态变化。STSpikeFuse 模块采用可学习的时序衰减机制，将 SNN 特征与语义表示进行自适应融合，这些语义表示由传统人工神经网络（ANN）主干生成。在变化检测数据集上的大量实验表明，SpikeAdapter 有效增强了时序感知能力与可解释性。
英文原文摘要 1306 字符
The temporal evolution patterns of surface spatial structures constitute a central concern within the field of intelligent remote sensing interpretation.However, constrained by the availability of only two temporal phases, modeling sparse spatio-temporal change processes to effectively interpret surface alterations remains a core challenge in intelligent remote sensing analysis. To address this, this paper proposes SpikeAdapter, a lightweight enhancement framework. This framework comprises Geo-Spike Interpolation (GSI-P), an spiking neural network (SNN) feature extractor, and the spatio-temporal fusion module STSpikeFuse. Inspired by the brain's perceptual response to new and fading stimuli, the core GSI-P module transforms bi-temporal radiometric differences into sparse spike sequences with time-to-first-spike characteristics.Then we use a feature extractor of SNN to capture dynamic variations of land-surface targets. The STSpkeFuse module employs a learnable temporal decay mechanism to adaptively fuse the SNN features with the semantic representations. This representations are generated by a traditional artificial neural network (ANN) backbone.Extensive experiments on change detection datasets demonstrate that SpikeAdapter effectively enhances temporal awareness and interpretability.