Matching and registration between aerial images and prestored road landmarks are critical techniques to enhance UAS (Unmanned Aerial System) navigation in the Global Positioning System (GPS)-denied urban environments.
Current registration processes typically consist of two separate stages of road extraction and road registration. These two-stage registration approaches are time-consuming and less robust to noise.
To that end, it has been investigated the problem of end-to-end Aerial-Road registration. Using deep learning, it has been developed a novel attention-based neural network architecture for Aerial-Road registration.
In this model, it has been constructed two-branch neural networks with shared weights to map two input images into a common embedding space. Besides, considering that road features are sparsely distributed in images, it has been incorporated a novel multibranch attention module to filter out false descriptor matches from the indiscriminative background in order to improve registration accuracy.
Finally, the results from extensive experiments show that compared with state-of-the-art approaches, the mean absolute errors of the approach in rotation angle and the translations in the x- and y-directions are reduced down by a factor of 1.24, 1.38, and 1.44, respectively. Furthermore, as a byproduct, the experimental results prove the feasibility of a neural network multitask learning approach to simultaneously achieve accurate Aerial-Road matching and registration, thus providing an efficient and accurate UAS geolocalization.
More info:
