Search

The rapid development of remote sensing technology has provided new sources of data for marine rescue and has made it possible to find and track survivors. Due to the requirement of tracking multiple survivors at the same time, multi-object tracking (MOT) has become the key subtask of marine rescue. However, there exists a significant gap between fine-grained objects in realistic marine rescue remote sensing data and the fine-grained object tracking capability of existing MOT technologies, which mainly focuses on coarse-grained object scenarios and fails to track fine-grained instances. Such a gap limits the practical application of MOT in realistic marine rescue remote sensing data, especially when rescue forces are limited. Given the promising fine-grained classification performance of recent text-guided methods, we delve into leveraging labels and attributes to narrow the gap between MOT and fine-grained maritime rescue. We propose a text-guided multi-class multi-object tracking (TG-MCMOT) method. To handle the problem raised by fine-grained classes, we design a multi-modal encoder by aligning external textual information with visual inputs. We use decoding information at different levels, simultaneously predicting the category, location, and identity embedding features of objects. Meanwhile, to improve the performance of small object detection, we also develop a data augmentation pipeline to generate pseudo-near-infrared images based on RGB images. Extensive experiments demonstrate that our TG-MCMOT not only performs well on typical metrics in the maritime rescue task (SeaDronesSee dataset), but it also effectively tracks open-set categories on the BURST dataset. Specifically, on the SeaDronesSee dataset, the Higher Order Tracking Accuracy (HOTA) reached a score of 58.8, and on the BURST test dataset, the HOTA score for the unknown class improved by 16.07 points. [ABSTRACT FROM AUTHOR]

Vegetation regulates the exchange of terrestrial carbon and water fluxes and connects the biosphere, hydrosphere, and atmosphere. Over the last four decades, vegetation greening has been observed worldwide using satellite technology. China has also experienced a notably widespread greening trend. However, the responsiveness of vegetation dynamics to elevated CO2 concentration, climate change, and human activities remains unclear. In this study, we attempted to explore the impact of natural (precipitation, air temperature), biogeochemical (CO2), and anthropogenic drivers (nighttime light, afforestation area) on changes in vegetation greenness in the Haihe River Basin (HRB) during 2002–2018 at the county-level. We further determined the major factors affecting the variation in satellite-derived normalized difference vegetation index (NDVI) from moderate resolution imaging spectroradiometer (MODIS) for each county. The results indicated that over 85% of the counties had a significantly increased NDVI trend, and the average linear trend of annual NDVI across the study region was 0.0037 per year. The largest contributor to the NDVI trend was CO2 (mean contribution 45%), followed by human activities (mean contribution of 27%). Additionally, afforestation was a pronounced driving force for NDVI changes in mountainous areas, resulting from ecosystem restoration efforts. Our findings emphasize the crucial role of CO2 fertilization in vegetation cover change, while considering CO2 concentration, climate change, and human activities, and shed light on the significant influences of afforestation programs on water resources, especially in mountainous areas. [ABSTRACT FROM AUTHOR]