Abstract:
Change detection is a fundamental task in remote sensing, supporting applications such as disaster response, land monitoring, infrastructure assessment, and security-related observation. Conventional change detection workflows generally depend on downlinking satellite imagery to ground stations, where preprocessing, analysis, interpretation, and product generation are performed. Although this ground-based paradigm has enabled reliable large-scale monitoring, it can introduce substantial latency between image acquisition and actionable information. This delay limits the value of satellite observations in time-critical scenarios where early alerts are essential. On-orbit change detection provides a promising alternative by enabling satellites to identify, prioritize, and report relevant surface changes directly onboard. This study reviews the opportunities and challenges of on-orbit change detection for remote sensing applications. We first examine the latency components of ground-based change detection, including observation lag, data lag, sensitivity lag, and production lag, and discuss how onboard processing may reduce or reshape these delays. We then synthesize traditional change detection methods, including image differencing, post-classification comparison, and time-series analysis, and evaluate how they can be adapted for resource-constrained spacecraft environments. Finally, we discuss key challenges for operational deployment, including lightweight processing, model updating, validation under lag-accuracy trade-offs, and change-category identification. The review shows that on-orbit change detection is unlikely to replace ground-based analysis entirely, but can provide earlier alerts, reduce unnecessary data downlink, and support more responsive Earth observation systems through staged and adaptive monitoring workflows. |