Answers are grounded in this paper and cite their sources.
In this paper, a synthetic aperture radar (SAR) non-coherent change detection (NC-CD) method based on the universal quality index (UQI) and gamma correction (GC) is proposed. The proposed method consists of two main steps: UQI-based coherence image generation, which jointly considers the local mean, variance, and covariance information between two SAR intensity images; and GC-based filtering after negative transformation to suppress residual speckle noise. On the Bern dataset, the proposed method achieved FA=0.677 %, κ=0.855, and PCD=99.32 %, reducing FA by 87.7 % and 82.1 % compared with DI and LRD, respectively. On the Ottawa dataset, it achieved FA=4.134 %, DR=86.18 %, κ=0.844, and PCD=95.87 %, reducing FA by 27.1 % and 15.5 % compared with DI and LRD, respectively. An additional sensitivity analysis of the UQI window size and GC order p, together with an ablation study, verified that the proposed framework effectively balanced changed-region preservation and speckle suppression.
No takes yet. Share an insight, caveat, or question.
Minseok Kang (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: