Search

Generative adversarial networks (or GANs) are a specific deep learning architecture often used for different usages, such as data generation or image-to-image translation. In recent years, this structure has gained increased popularity and has been used in different fields. One area of expertise currently in vogue is the use of GANs to produce synthetic aperture radar (SAR) data, and especially expand training datasets for SAR automatic target recognition (ATR). In effect, the complex SAR image formation makes these kind of data rich in information, leading to the use of deep networks in deep learning-based methods. Yet, deep networks also require sufficient data for training. However, contrary to optical images, we generally do not have a substantial number of available SAR images because of their acquisition and labelling cost; GANs are then an interesting tool. Concurrently, how to improve explainability for SAR ATR deep neural networks and how to make their reasoning more transparent have been increasingly explored as model opacity deteriorates trust of users. This paper aims at reviewing how GANs are used with SAR images, but also giving perspectives on how GANs could be used to improve interpretability and explainability of SAR classifiers. [ABSTRACT FROM AUTHOR]

Understanding the decision-making process of deep learning networks is a key challenge that has rarely been investigated for synthetic aperture radar (SAR) images. In this article, a set of new analytical tools is proposed and applied to a convolutional neural network (CNN) handling automatic target recognition on two SAR datasets containing military targets. First, an analysis of the respective influence of target, shadow, and background areas on classification performance is carried out. The shadow appears to be the least used portion of the image affecting the decision process, compared to the target and clutter, respectively. Second, the location of the most influential features is determined with classification maps obtained by systematically hiding specific target parts and registering the associated classification rate relative to the images to be classified. The location of the image areas without which classification fails is target type and orientation specific. Nonetheless, a strong contribution of specific parts of the target, such as the target top and the areas facing the radar, is noticed. Finally, results show that features are increasingly activated along the CNN depth according to the target type and its orientation, even though target orientation is absent from the loss function. [ABSTRACT FROM AUTHOR]

This paper addresses the estimation of the height of a point scatterer over a sea surface via multipath exploitation for a High Range Resolution radar that is using pulse range compression, such as Synthetic Aperture Radars. We first focus our attention on the physical model, in particular on the specular/diffuse reflection coefficients, this coefficients being derived from the empirical Miller Brown and Vegh model. The gravity waves are also simulated since they modify the acquisition geometry such as the local grazing angle. Secondly, the signal model is derived, thus allowing an easy derivation of the time delays (direct echo and replicas), these time delays being converted into a height estimation for possible automatic ship recognition applications. Our algorithm is a non-conventional radar signal processing, in other words it uses the backscattered pulse over before range compression and demodulation. The aim of the paper is to understand for which radar and sea parameters, as well as acquisition scenes, it is possible to extract the scatterer height information using the multipath of the backscattered electromagnetic wave. [ABSTRACT FROM AUTHOR]