VL-MFL: UAV Visual Localization Based on Multisource Image Feature Learning

Obtaining the Earth-fixed coordinates is a fundamental requirement for long-distance unmanned aerial vehicle (UAV) flight. Global navigation satellite systems (GNSSs) are the most common location model, but their signals are susceptible to interference from obstacles and complex electromagnetic environments. To solve this issue, a visual localization framework based on multisource image feature learning (VL-MFL) is proposed. In the proposed framework, the UAV is located by mapping airborne images to the satellite images with absolute coordinate positions. First, for the heterogeneity issues caused by different imaging environments of drone and satellite images, a lightweight Siamese network based on 3-D attention mechanism is proposed to extract consistent features from multisource images. Second, to overcome the problem of inaccurate localization caused by the large receptive field of traditional convolutional neural networks, the cell-divided strategy is imported to strengthen the position mapping relationship of multisource images features. Finally, based on similarity measurement, a confidence evaluation mechanism is established and a search region prediction method is proposed, which effectively improved the accuracy and efficiency in matching localization. To evaluate the location performance of the proposed framework, several related methods are compared and analyzed in detail. The results on the real-world datasets indicate that the proposed method has achieved outstanding location accuracy and real-time performance.