New method improves accuracy and speed of satellite change detection
Temporally Ordered Region-Token Mamba with Logit-Space Diffusion for Remote Sensing Change Detection
Computer Vision and Pattern Recognition
Summary
Detecting changes in satellite images over time is tricky because images are large and changes can be subtle. The authors developed a new technique called BMD-CD that organizes image data into time-based groups and uses a special process to focus on true changes while ignoring small variations. This method works well at finding and outlining changed areas accurately and runs quickly on typical image sizes. Tests on several well-known satellite image datasets show it performs better than previous approaches and can even work on new datasets without extra training.
What this means in practice
- •For satellite image analysts: Process high-resolution satellite image pairs to accurately identify and outline changed regions with improved speed and boundary precision.
- •For environmental monitoring teams: Monitor land use changes by detecting precise modifications from satellite images more reliably across diverse geographic areas using zero-shot transfer.
Authors
Anuvab Sen, Maneet Chatterjee, Aparup Ghosh, Udayon Sen, Arnav Aditya, Yixin Zhang
Abstract
Remote sensing change detection requires both global reasoning across bitemporal images and precise localization of changed regions. However, dense attention is computationally expensive for high-resolution imagery, while conventional feature fusion and coarse decoding may inadequately separate genuine changes from appearance variations or preserve object boundaries. We present Bitemporal Mamba-Diffusion for Change Detection (BMD-CD), which combines temporally structured state-space modeling with logit-space diffusion refinement. BMD-CD converts deep bitemporal features into region tokens and arranges them in explicit temporal partitions before bidirectional state-space propagation. Its Bitemporal Ordered Mamba Operator enables long-range cross-temporal interaction with linear sequence complexity, while Orthogonal Feature Disentanglement forms a change-oriented output and a complementary rotated output using learned pairwise rotations and unchanged-region consistency. Multiscale decoding then produces coarse change logits, which are refined through a five-step Conditional Diffusion Decoder operating directly in logit space. Experiments on LEVIR-CD, WHU-CD, DSIFN-CD, CDD, and S2Looking demonstrate strong performance across diverse change-detection settings. BMD-CD achieves F1 scores of 93.7%, 96.0%, 97.8%, and 99.0% on the four standard benchmarks and improves 3-pixel Boundary-F1 to 87.7% and 91.4% on LEVIR-CD and WHU-CD, respectively. The full model requires 32.09 GFLOPs and 47 ms per 256 x 256 image pair, while also showing zero-shot transfer to ValaisCD and B-FLAIR-test. Our code is available at https://github.com/Aparup2139/Public_WACV/