Search Results (1,764)

Unmanned Aerial Vehicle Ad Hoc Networks (UAVANETs) are characterized by highly dynamic topology changes and unstable link conditions, which necessitate deep collaboration between transport-layer congestion control and network-layer routing decisions to ensure service quality. However, the existing layered architecture of Software-Defined Networking (SDN) results in a significant separation between routing information and congestion control mechanisms, rendering traditional protocols ineffective in handling severe performance fluctuations caused by highly dynamic route switching. The significant disconnect between network-layer route planning and transport-layer congestion control strategies in Software-Defined Unmanned Aerial Vehicle Ad Hoc Networks (SD-UAVANETs) leads to degraded transmission performance of BBR (Bottleneck Bandwidth and Round-trip propagation time) under high-dynamic route switching scenarios. As such, this paper proposes an in-band network telemetry (INT)-based cross-layer optimization scheme for BBR, named SDN-BBR. Firstly, a lightweight real-time route switching detection mechanism based on INT is designed. Secondly, a QoS inequality model before and after path switching is established, deriving the critical bandwidth of the new path and integrating it into the BBR algorithm to accelerate convergence and avoid congestion. Finally, the BBR state machine is redesigned to achieve cross-layer information fusion and coordinated control, thereby optimizing transmission performance. Experimental results show that the proposed scheme reduces convergence time by 69.8% and increases throughput by 73.9% in low-bandwidth to high-bandwidth switching scenarios; decreases packet loss rate by 86.8% and reduces delay by 8.3% in high-bandwidth to low-bandwidth switching scenarios; and improves throughput by 12.3%, lowers packet loss rate by 21%, and reduces delay by 7.9% in multi-traffic flow concurrent scenarios. The scheme significantly enhances the transmission performance of BBR in highly dynamic routing environments of SD-UAVANET. Full article

Dissemination and conservation of cultural heritage have been challenged by continued accessibility in museums, where traditional information delivery systems are at times ineffective in terms if interaction with visitors. The current paper investigates RumiArt IA, a mobile application, to identify cultural objects in real-time, remaining fully in the scope of this line of research without relying on internet connectivity. The system, which is developed based on the Rumiñahui Museum and Cultural Center, Ecuador, uses transfer learning in the MobileNetV2 architecture with INT8 post-training quantization to identify 21 cultural artifacts spread across six thematic rooms. The experiment involved building a dataset of 36,000 images under diverse lighting conditions, viewing angles, and distances; furthermore, artificial transformations were explicitly crafted to simulate real museum conditions such as glass reflections and non-frontal capture angles. Quantization was used to reduce each model to 775 KB as compared with the 2.4 MB, with accuracy loss not reaching more than 0.5 percent (DKL < 0.05). Assessment of 9450 validation images yielded a general accuracy of 92.2%, with an inference time of 63 ms on current devices with a high throughput and 215 ms on mid-range hardware from 2020. Practical validation involving 50 visitors of the museum showed a success rate of 93.7%, with average user satisfaction at 8.5/10 and 87%, indicating they would recommend the application. An in-depth error study of the most difficult room (88.3% accuracy) indicated that 47% of the errors were due to the angles of the camera, which blocked out distinguishing features, and 22% were caused by display case reflections and the shadows of the visitors. These results indicate that end-to-end machine learning can provide consistent cultural heritage recognition in resource-constrained settings but its efficiency is susceptible to physical capture factors that cannot be resolved by data augmentation. Offline mode and low memory footprint (less than 90 MB when loaded on six models) of the system are especially relevant to application in situations where there is no guarantee of cloud connectivity. Full article

Soil salinization is one of the primary forms of land degradation in arid and semi-arid regions, severely constraining agricultural production in Xinjiang’s oases. Unmanned aerial vehicle (UAV) imagery provides an effective means for precise monitoring of soil salinization, with image spatial resolution being a key factor affecting assessment accuracy. However, traditional single-scale remote sensing monitoring methods rely solely on spectral and textural features at the leaf scale (0.1 m resolution captures leaf-scale characteristics), neglecting the contribution of multi-scale features (single-row canopy scale and single-membrane-covered area scale (6-row crop canopy)) to soil salinity. For instance, 0.5–1 m reflects single-row canopy scale, while 2 m reflects single-membrane-covered area scale. Therefore, this study developed a multi-scale UAV imagery and machine learning framework to enhance soil electrical conductivity prediction accuracy. This study focuses on oasis cotton fields in Shaya County, Xinjiang. Based on UAV multispectral imagery, we resampled data to generate eight datasets at different spatial resolutions: 0.1, 0.5, 1, 1.5, 2, 2.5, 5, and 10 m. For each resolution, we calculated 21 spectral indices and 48 texture features to construct a feature set. At both single and multispatial scales, spectral indices, texture features, and their spectral-texture fusion features were constructed. Combining these with Backpropagation Neural Network (BPNN), Random Forest Regression (RFR), and Extreme Gradient Boosting (XGBoost) models, a soil EC estimation framework was developed. The impact of three feature combination schemes on cotton field soil conductivity estimation using single-scale UAV imagery was compared. The accuracy of soil EC estimation for cotton fields was compared between multi-spatial scale and single-scale UAV image features. The optimal combination strategy for a multi-spatial scale and multiple features was determined. Results indicate that combining spectral and texture features yields the highest estimation accuracy for cotton field soil electrical conductivity in single-scale analysis. Multi-spatial scale image features outperform single-scale image features in estimating cotton field soil electrical conductivity accuracy. By comparing different feature combinations, when integrating 0.5 m spatial-scale spectra (S1, EVI, DVI, NDVI, Int1, SI) with 0.1 m texture features (RE1_ent, R_cor, RE1_cor, G_hom, B_mea, R_con, NIR_con), the XGBoost model achieved the optimal prediction accuracy ($R^2$ = 0.693, RMSE = 0.515 dS/m), outperforming the methods using multiple features at a single scale. This study developed a novel multi-scale image feature fusion technique to construct a machine learning model. This method describes the image characteristics of soil electrical conductivity at different geographical scales, providing a reference approach for the rapid and accurate prediction of soil electrical conductivity in arid regions. Full article

Rheumatoid arthritis (RA) is a chronic autoimmune disease characterised by persistent joint inflammation and systemic immune dysregulation. While bone marrow activation has been linked to RA pathogenesis, direct access to bone marrow tissue for progenitor analysis remains limited by ethical and technical constraints. Analysis of progenitor cells in peripheral blood can serve as a surrogate reflecting bone marrow activation. In this study, we analysed peripheral blood cells from 12 RA patients and 9 healthy controls using high-dimensional spectral flow cytometry with a nine-marker panel (CD45, CD31, CD235, CD133, CD34, CD105, CD271, CD90, PDPN). Flow Self-Organizing Map (FlowSOM) clustering identified 20 distinct cell populations. Additionally, a complementary flow cytometry panel was used to assess CD31 expression on immune subsets in peripheral mononuclear cells (PBMCs) from 9 RA and 9 healthy donors of this cohort. RA patients showed increased CD45⁺CD31⁻ immune cells, but not their putative progenitors. Conversely, putative CD45⁺CD31^int progenitors and CD45⁺CD31^int mature cells were reduced, along with CD31 expression on T cells. Levels of CD235a⁺ putative erythroid precursors and CD45⁺CD31⁺ progenitors were significantly increased in RA patients. Three putative stromal cell populations were detected in circulation. Together, these findings reveal expanded erythroid precursor populations and reduced CD31 expression on T cells in RA. Our data underscore broad systemic alterations in cellular homeostasis in RA patients. In conclusion, our results suggest that the loss of CD31 expression on immune cell precursors plays a role in age-associated immune remodelling and immune activation in RA and provides the rationale for further studies on erythroblast differentiation and the functional role of erythroblasts in chronic inflammation. Full article

Despite being a biodiversity hotspot, the Republic of Seychelles faces a critical challenge with an estimated 90% of its food imported. This dependency exposes the country to global supply disruptions and climate-related risks, while pressure on protected ecosystems continues to rise. In response, the Erasmus+ Capacity Building Higher Education GreenTraINT project (Green Training INTernational Program for agriculture, livestock farming, and conservation), co-funded by the European Union (2024–2026), aims to strengthen local expertise in sustainable agriculture, livestock farming, and biodiversity conservation. Through a transnational partnership involving European and Seychellois universities and institutions, GreenTraINT is co-designing innovative higher education modules tailored to the island’s priorities in agriculture, livestock, and biodiversity conservation. This paper focuses on a detailed needs analysis conducted in early 2025 across a diverse group of 84 stakeholders, including students, educators, NGOs, and professionals. The findings reveal a strong demand for applied training in sustainable food systems and biodiversity conservation, blended teaching methods, and programs that bridge theory with hands-on skills. Inspired by other Erasmus+ projects such as NETCHEM and SPARKLE, GreenTraINT adopts a multi-stakeholder, needs-driven approach that aligns international academic expertise with local development goals. As a key milestone, a Summer School in 2026 will pilot the newly developed modules. In the long term, GreenTraINT seeks to leave a lasting legacy by integrating its curriculum into national education pathways, thereby contributing to food security and environmental resilience. With less than four years remaining to achieve the 2030 Agenda targets, the project positions higher education reform as a strategic accelerator for SDG implementation in small island developing states (SIDS). By linking curriculum innovation to measurable sustainability priorities, GreenTraINT helps narrow the SDG implementation gap in vulnerable island contexts. The project offers a model for international collaboration in higher education for sustainability in SIDS. Full article