Search

Loop closure detection is an important component of the Simultaneous Localization and Mapping (SLAM) algorithm, which is utilized in environmental sensing. It helps to reduce drift errors during long-term operation, improving the accuracy and robustness of localization. Such improvements are sorely needed, as conventional visual-based loop detection algorithms are greatly affected by significant changes in viewpoint and lighting conditions. In this paper, we present a semantic spatial structure-based loop detection algorithm. In place of feature points, robust semantic features are used to cope with the variation in the viewpoint. In consideration of the semantic features, which are region-based, we provide a corresponding matching algorithm. Constraints on semantic information and spatial structure are used to determine the existence of loop-back. A multi-stage pipeline framework is proposed to systematically leverage semantic information at different levels, enabling efficient filtering of potential loop closure candidates. To validate the effectiveness of our algorithm, we conducted experiments using the uHumans2 dataset. Our results demonstrate that, even when there are significant changes in viewpoint, the algorithm exhibits superior robustness compared to that of traditional loop detection methods. [ABSTRACT FROM AUTHOR]

As the core area of human activities and economic development in the Xinjiang Autonomous Region, the hilly oasis zone of Xinjiang directly affects the regional sustainable development and stability of the ecosystem. Understanding the effects of different geomorphic types on vegetation distribution is crucial for maintaining vegetation growth and development, especially the improvement in the terrestrial ecological environment in arid areas under the background of climate change. However, there are few studies on the effect of spatial differences in detailed geomorphic types on vegetation distribution patterns. Therefore, this paper divides the Xinjiang hilly oasis zone into six geomorphologic level zones and innovatively investigates the influence of detailed geomorphologic types on the spatial distribution of vegetation and vegetation cover. Further, the area proportion of detailed landform types corresponding to different vegetation coverage in each geomorphic area was quantitatively calculated. Finally, the Geodetector method was used to detect the drivers of interactions between vegetation and the environment. The findings are shown as follows: (1) In the same climate zone, the spatial differentiation of landforms has a great influence on the vegetation distribution, manifesting as the significantly different vegetation distribution in different landform types. Grassland is the main vegetation type in the erosion and denudation of Nakayama; cultivated vegetation and meadows have a larger coverage in the alluvial flood plain and alluvial plain; and the distribution of vegetation in the Tianshan economic zone is characterized by obvious vertical zoning with the geomorphology. (2) The landform type and morphological types are the strongest driving factors for vegetation coverage with q values of 0.433 and 0.295, respectively, which effectually fill the gap caused by only using two terrain indicators, slope and elevation, to study the relationship between landforms and vegetation. (3) In addition, the improved nonlinear interaction resulting from the double factor of landform type and slope is 0.486, which has a stronger control on vegetation coverage than the single factor of landform type. These findings are conducive to enhancing the supply services of vegetation to the ecosystem in arid areas as well as providing important scientific guidance for the construction of ecological civilization and sustainable development in Xinjiang. [ABSTRACT FROM AUTHOR]