Opportunistic Bistatic SAR Image Classification Using Sub-aperture Decomposition

Spaceborne synthetic aperture radar (SAR) can provide accurate images of the ocean surface roughness dayor-night in nearly all weather conditions, being an unique asset for many geophysical applications. Considering the huge amount of data daily acquired by satellites, automated techniques for physical features extraction are needed. Even if supervised deep learning methods attain state-of-the-art results, they require great amount of labeled data, which are difficult and excessively expensive to acquire for ocean SAR imagery. To this end, we use the subaperture decomposition (SD) algorithm to enhance the unsupervised learning retrieval on the ocean surface, empowering ocean researchers to search into large ocean databases. We empirically prove that SD improve the retrieval precision with over 20% for an unsupervised transformer auto-encoder network. Moreover, we show that SD brings important performance boost when Doppler centroid images are used as input data, leading the way to new unsupervised physics guided retrieval algorithms.

show abstract

Section: A Subaperture Decompositionmentioning

confidence: 99%

“…The SD algorithm is widely used for SAR imagery [16], [17], [18], [19], [20], [15]. The method was combined with both classical signal processing algorithms [17], [18], [19] and deep learning techniques [16], [15]. In [17] the SD is proposed for the ship detection task, while in [18] it is used for target characterization.…”

Section: A Subaperture Decompositionmentioning

confidence: 99%