Search Results (3)

Remote sensing small object detection remains a challenging task due to limited feature representation and interference from complex backgrounds. Existing methods that rely exclusively on either visible or infrared modalities often fail to achieve both accuracy and robustness in detection. Effectively integrating cross-modal information to enhance detection performance remains a critical challenge. To address this issue, we propose a novel Multispectral Sparse Cross-Attention Guided Mamba Network (MSCGMN) for small object detection in remote sensing. The proposed MSCGMN architecture comprises three key components: Multispectral Sparse Cross-Attention Guidance Module (MSCAG), Dynamic Grouped Mamba Block (DGMB), and Gated Enhanced Attention Module (GEAM). Specifically, the MSCAG module selectively fuses RGB and infrared (IR) features using sparse cross-modal attention, effectively capturing complementary information across modalities while suppressing redundancy. The DGMB introduces a dynamic grouping strategy to improve the computational efficiency of Mamba, enabling effective global context modeling. In remote sensing images, small objects occupy limited areas, making it difficult to capture their critical features. We design the GEAM module to enhance both global and local feature representations for small object detection. Experiments on the VEDAI and DroneVehicle datasets show that MSCGMN achieves mAP50 scores of 83.9% and 84.4%, outperforming existing state-of-the-art methods and demonstrating strong competitiveness in small object detection tasks. Full article

In the Mediterranean, anthropogenic pressures (specifically those involving nutrient loads) have been progressively moved to deeper off-shore areas to meet current policies dealing with the protection of marine biodiversity (e.g., European Directives). However, conservation efforts devoted to protecting Posidonia oceanica and other vulnerable marine habitats against anthropogenic pressures have dedicated very little attention to the deepest areas of these habitats. We studied the remote influence of off-shore nutrient discharge on the physiology and structure of deep P. oceanica meadows located nearest to an urban sewage outfall (WW; 1 km) and an aquaculture facility (FF; 2.5 km). Light reduction and elevated external nutrient availability (as indicated by high δ¹⁵N, total N and P content and N uptake rates of seagrass tissues) were consistent with physiological responses to light and nutrient stress. This was particularly evident in the sites located up to 2.5 km from the WW source, where carbon budget imbalances and structural alterations were more evident. These results provide evidence that anthropogenic nutrient inputs can surpass critical thresholds for the species, even in off-shore waters at distances within the km scale. Therefore, the critical distances between this priority habitat and nutrient discharge points have been underestimated and should be corrected to achieve a good conservation status. Full article

For some time, there has been a slow but gradual rise in the average temperature of the entire globe, a “global warming”, in fact, the result of human and natural processes that have been producing this phenomenon for decades. Since they are not directly perceived by individuals, these processes and their effects have been ignored for a long time, or at least not considered to be immediately harmful and dangerous. Global warming does not depend so much on solar radiation as it does on the greenhouse effect deriving from the continuous emission, by human activities and natural events, of greenhouse gases that accumulate in the atmosphere and form a barrier to the dispersion of heat produced by solar radiation. A good number of models exists to explain how global warming is produced, which are technical in nature and consider the production of greenhouse gases as the most important cause; however, they do not always analyze and justify the reasons for such emissions. Following the logic, language and methods of Senge’s systems thinking, the paper aims to present a general model, the GEAM—qualitative in nature, but rational and coherent—for highlighting the interacting factors that give rise to and maintain global warming. This model constitutes a reference framework to identify possible “strategic areas” within which to identify man-made “artificial” and “natural” factors that can control the phenomenon and to order the countless ideas and interventions that different nations carry out individually to control global warming. The model presented is qualitative in nature and does not allow immediate calculations or forecasts to be performed. However, it could guide in-depth scientific research in generating accurate forecasts and simulation using the tools of systems dynamics. In conclusion, understanding how global warming is created and if and how it could be controlled is the aim of this work. Finally, I want to note that the purpose of this work is not to analyze the technical aspects of the phenomenon of global warming, or to deepen the measures and actions to contrast it, but to provide a “general model of description and understanding” of the phenomenon using logic and language of the Systems Thinking Approach (according to Peter’s Senge and Piero Mella), with the aim of highlighting three fundamental strategic areas for countering the phenomenon and four uncontrollable phenomena, triggered by global warming itself, which can make strategic control difficult. Furthermore, I highlight the role played by the world population, understood both quantitatively as a number and qualitatively as a level of economic development, in the production of global warming. Lastly, I observe how the different strategic actions that nations can, indeed must, implement to stem global warming are systematically interconnected and interacting; however, these interactions can produce unknown effects and consequences that must be carefully researched and evaluated—encouraged, if positive, reduced or eliminated if negative. Full article