MambaMPD improves marine pollution detection from satellite images
MambaMPD: A Mamba-Driven Segmentation Framework for Marine Pollution Detection from Remote Sensing Imagery
Computer Vision and Pattern Recognition
Summary
Detecting pollution in oceans from satellite pictures is hard because pollutants often look like the surrounding sea and have unclear shapes. The paper presents MambaMPD, a new computer method that uses a special type of AI model called Mamba. It includes techniques to better recognize the rough edges and low-contrast parts of pollution, making it easier to spot oil spills and other contaminants from space. Tests show it works better and faster than earlier methods on standard marine pollution datasets.
What this means in practice
- •For environmental monitoring teams: Detect and map marine pollution from satellite images with improved accuracy and efficiency using MambaMPD.
- •For coastal cleanup organizations: Use detailed pollution boundary maps from MambaMPD to better plan and target cleanup efforts.
Authors
Shuaiyu Chen, Wei Han, Peng Ren, Chunbo Luo, Zeyu Fu
Abstract
Accurate marine pollution detection (MPD) is essential for protecting coastal ecosystems and marine biodiversity. Vision Mamba models have shown promise in remote-sensing semantic segmentation by efficiently capturing long-range dependencies and global context, yet their potential for MPD remains underexplored. MPD is particularly challenging because of low signal-to-noise ratios, fragmented pollution patterns, and indistinct boundaries caused by the visual similarity between pollutants and the surrounding sea. To address these issues, we propose MambaMPD, an enhanced Mamba-based framework incorporating two complementary structural priors: Frequency-Aware Augmentation (FAA) and multi-scale Edge-Guided Attention (EGA). FAA integrates wavelet transforms into the encoder to decompose features into multi-scale frequency subbands, enabling the model to capture low-frequency contextual semantics and high-frequency structural details needed to identify small, low-contrast, and irregular pollution patterns. EGA adaptively fuses hierarchical, Laplacian-derived boundary cues with deep semantic representations, refining encoder features before decoding to sharpen boundaries and reduce ambiguity in visually confusing, spatially fragmented scenes. Together, these modules improve sensitivity to subtle pollution signals while preserving fine boundary structures. A U-Net-style decoder with squeeze-and-excitation attention and deep supervision progressively restores and refines semantic and spatial information across scales. Extensive experiments on two benchmark MPD datasets show that MambaMPD achieves higher mIoU than competing methods while requiring substantially less computation than foundation-model-based approaches. On MADOS, it improves F1 by 3.6% over OSDMamba; on M4D, it raises Oil Spill IoU by 6.82% over TransOilSeg.