Search

A dust devil is an important part of the Martian climate system, which can help us better understand scientific questions of the climate, surface–atmosphere interactions, aeolian processes, and regolith on Mars. Therefore, the automatic detection of dust devils from Mars orbiter images is becoming increasingly important for the scientific study and the planning of future robotic and manned missions. To improve the generalization, detection efficiency, and accuracy of the traditional approach in automatically detecting dust devils, we made several modifications to the faster region-based convolution neural network. Based on the characteristics of the dust devil, we proposed a Martian dust devil detection network (MDDD Net). The network uses the feature pyramid network to obtain a feature fusion map with rich location information and semantic information. The k-means++ algorithm is used to design reasonable anchor boxes to adapt to vary sized dust devils. The region of interest align unit is introduced to eliminate the mapping deviation between the feature map and the original image. Finally, the soft nonmaximum suppression algorithm is used to complete the screening of the bounding box. It can reduce missing detections caused by the overlapping between adjacent dust devil bounding boxes in the same image. The average precision and recall of MDDD Net on the dust devil dataset built in this article reach 90.1% and 96.5%, respectively.

Abstract Mars' surface is characterized by a weathered layer of regolith and exposed rock exposures that are the results of long‐term geological processes. The Mars Rover Penetrating Radar (RoPeR) on board the Zhurong rover of China's first Mars mission (Tianwen‐1) has been investigating the fine structure and dielectric properties of the martian regolith in the southern Utopia Planitia. The permittivity of the regolith within 5 m of the landing zone is 3.6−2.0+3.1, and the average loss tangent is 0.0060 ± 0.0002, 0.0087 ± 0.0002, and 0.0114 ± 0.0002 using the geometric compensation of R2, R3, R4, respectively. The permittivity distribution map has been derived to show permittivity varying with depth and location. The high dispersion of both permittivity and loss tangent values along the traverse path indicates relatively heterogeneous material distribution on the landing site compared to an airless body such as the Moon.