Target detection is a cornerstone task in hyperspectral image processing but faces significant challenges due to domain gaps. While statistical detectors like Constrained Energy Minimization (CEM) and Adaptive Cosine Estimator (ACE) are not prone to learned biases, in practice they still suffer from mismatches between the reference target spectrum and the spectral characteristics of the target in the test scene. We propose Test-time Adaptive Spectrum Refinement (TASR), a novel framework addressing this problem. TASR operates in an interpretable, lightweight, data-efficient manner, requiring only a single labeled source image of the target material. At test time, TASR dynamically refines the target spectrum to better align with the spectral properties of the test scene. This adaptive refinement enables detectors to effectively handle data with spectral variations, bridging the gap between the source and test spectra. To validate TASR, we conduct extensive experiments on established benchmarks and introduce a new dataset—ShadySunnyDiffuse (SSD)—which explicitly tests detector robustness to naturally occurring illumination changes. We further demonstrate the method’s versatility by applying it to camouflage detection and show compatibility with multiple statistical detectors. Our results establish TASR as a state-of-the-art approach in domain-adaptive hyperspectral target detection and target spectrum management.

Robust hyperspectral target detection depends on a reliable reference spectrum that characterizes the material of interest. However, this spectrum is typically extracted from labeled source images that differ from the test scene in illumination, background, or atmospheric conditions-introducing domain shifts that degrade performance. While recent work has shown that discrete genetic algorithms can refine a single-source spectrum to improve cross-scene generalization, spectrum adaptation remains an underexplored avenue in hyperspectral domain adaptation. We present Multi-Source Adaptive Constrained Energy Minimization (MACEM), which, to the best of our knowledge, is the first real-time method that explicitly optimizes a reference spectrum to bridge spectral domain shifts while leveraging multiple labeled sources. MACEM enhances the robustness of Constrained Energy Minimization (CEM) across diverse and challenging test environments. It is few-shot, deterministic, and transparent, making it well-suited for time-critical and mission-critical applications such as UAV-based threat detection. Across various cross-domain benchmarks, MACEM not only outperforms naive transfer baselines, but also achieves substantial gains over the previous state-of-the-art spectrum refinement approach, advancing the standard for robust hyperspectral target detection using statistical methods.