Summary
Recognizing objects in radar images is hard because radar pictures look very different from regular photos and there is little labeled radar data. The researchers explored using models trained on large collections of optical images, like normal photos, to help classify radar images. They created a way for a radar image classifier to learn from prototypes formed by an optical model without needing exact matching pairs of radar and optical images. Their method improved radar classification accuracy and showed clearer grouping of object types. This approach suggests that knowledge from visual models of regular photos can help improve recognition in radar images, even though they are very different types of images.
Synthetic Aperture Radar (SAR)Automatic Target Recognition (ATR)Electro-optical (EO) imageryVision foundation modelsDINOv3Cross-modal learningPrototypesImage classificationt-SNE visualizationDomain gap
Authors
Lucas Hirsch, James R. Hopgood, Javid Khan, Yoann Altmann, Mike E. Davies
Abstract
Synthetic Aperture Radar (SAR) is an important modality in a wide range of imaging applications due to its versatile, long range and near all weather operating capabilities. However, Automatic Target Recognition (ATR) remains a challenging problem due to limited labelled data, the strong speckle in SAR images and the significant domain gap between SAR and more abundant optical imagery. In contrast, electro-optical (EO) imagery benefits from massive datasets, clearer visual structure and powerful foundation models. In this work, we investigate how vision foundation models trained on optical data can provide class level supervision for SAR classification. We propose a cross-modal EO to SAR prototype alignment framework in which a frozen EO encoder, based on a DINOv3 vision foundation model, is used to construct class level optical prototypes without requiring strict EO/SAR pairs. A SAR model is then trained to classify SAR images while aligning its embeddings to the corresponding EO class prototype. At inference time, the SAR model operates independently, without access to optical imagery. We evaluate our approach on the UNICORNv2 dataset, an EO and SAR dataset of civilian vehicles with heavily speckled images and severe class imbalance. EO prototype alignment improves SAR classification accuracy over frozen DINOv3, SAR only finetuning and unpaired distribution alignment baselines, and t-SNE visualizations provide qualitative evidence of clearer separation among classes in the trained SAR embedding space. These results suggest that optical vision foundation models, despite being trained on visible spectrum imagery, provide transferable information for SAR image classification, offering a practical method for using large scale pretrained vision foundation models across challenging sensing modalities.