RESUMO
Nowadays, the focus on few-shot object detection (FSOD) is fueled by limited remote sensing data availability. In view of various challenges posed by remote sensing images (RSIs) and FSOD, we propose a meta-learning-based Balanced Few-Shot Object Detector (B-FSDet), built upon YOLOv9 (GELAN-C version). Firstly, addressing the problem of incompletely annotated objects that potentially breaks the balance of the few-shot principle, we propose a straightforward yet efficient data clearing strategy, which ensures balanced input of each category. Additionally, considering the significant variance fluctuations in output feature vectors from the support set that lead to reduced effectiveness in accurately representing object information for each class, we propose a stationary feature extraction module and corresponding stationary and fast prediction method, forming a stationary meta-learning mode. In the end, in consideration of the issue of minimal inter-class differences in RSIs, we propose inter-class discrimination support loss based on the stationary meta-learning mode to ensure the information provided for each class from the support set is balanced and easier to distinguish. Our proposed detector's performance is evaluated on the DIOR and NWPU VHR-10.v2 datasets, and comparative analysis with state-of-the-art detectors reveals promising performance.
RESUMO
Vast amounts of monitoring data can be obtained through various optical sensors, and mask detection based on deep learning integrates neural science into a variety of applications in everyday life. However, mask detection poses technical challenges such as small targets, complex scenes, and occlusions, which necessitate high accuracy and robustness in multi-scene target detection networks. Considering that multi-scale features can increase the receptive field and attention mechanism can improve the detection effect of small targets, we propose the YOLO-MSM network based on the multi-scale residual (MSR) block, multi-scale residual cascaded channel-spatial attention (MSR-CCSA) block, enhanced residual CCSA (ER-CCSA) block, and enhanced residual PCSA (ER-PCSA) block. Considering the performance and parameters, we use YOLOv5 as the baseline network. Firstly, for the MSR block, we construct hierarchical residual connections in the residual blocks to extract multi-scale features and obtain finer features. Secondly, to realize the joint attention function of channel and space, both the CCSA block and PCSA block are adopted. In addition, we construct a new dataset named Multi-Scene-Mask, which contains various scenes, crowd densities, and mask types. Experiments on the dataset show that YOLO-MSM achieves an average precision of 97.51%, showing better performance than other detection networks. Compared with the baseline network, the mAP value of YOLO-MSM is increased by 3.46%. Moreover, we propose a module generalization improvement strategy (GIS) by training YOLO-MSM on the dataset augmented with white Gaussian addition noise to improve the generalization ability of the network. The test results verify that GIS can greatly improve the generalization of the network and YOLO-MSM has stronger generalization ability than the baseline.
RESUMO
OBJECTIVES: We aim to analyse the effects of government subsidies on residents' health and healthcare expenditure from the perspectives of supply and demand. DATA AND METHODS: According to the regional division adopted in the data query system of the National Bureau of Statistics, this study divides 31 provinces and cities into three regions: eastern, central, and western. The data used are from public databases, such as the "China Statistical Yearbook," "China Health Statistical Yearbook," and "Government Final Account Report". In this study, mathematical model derivation is used to construct a fixed effects model, and an empirical study based on cross-sectional data and general linear regression is conducted. To prevent endogeneity issues, this study introduces instrumental variables and uses 2SLS regression to further analyse the output results. RESULTS: For every 1% increase in supplementary funding on the supply side, the perinatal mortality rate decreases by 1.765%, while for every 1% increase in financial compensation on the demand side, per capita outpatient expenses increase by 0.225% and per capita hospitalization expenses increase by 0.196%. Regarding medical resources, for every 1% increase in the number of beds per 1,000 people, per capita hospitalization expenses decrease by 0.099%. In the central and eastern regions, where economic levels are higher, supply-side government funding is more effective than demand-side funding. In contrast, demand-side funding is more effective in the western region. CONCLUSIONS: The roles of multiple influencing factors and significant regional heterogeneity are clarified. Increasing financial compensation to providers positively impacts perinatal mortality but leads to higher per capita outpatient and hospital expenditures. Finally, this study provides targeted policy recommendations and solid theoretical support for policymakers.
RESUMO
Microbiome manipulation is gaining fresh attention as a way to mitigate diseases in aquaculture. The commercially farmed seaweed Saccharina japonica suffers from a bacterial-induced bleaching disease, which has major implications for the reliable supply of healthy sporelings. Here, we identify a beneficial bacterium, Vibrio alginolyticus X-2 that significantly reduces the risk of bleaching disease. By combining infection assays and multi-omic analyses, we provide evidence to suggest that the underlying protective mechanisms of V. alginolyticus X-2 involve maintaining epibacterial communities, increasing the gene expression of S. japonica related to immune and stress protection pathways, and stimulating betaine concentrations in S. japonica holobionts. Thus, V. alginolyticus X-2 can elicit a suite of microbial and host responses to mitigate the bleaching disease. Our study provides insights into disease control in farmed S. japonica through the application of beneficial bacteria. IMPORTANCE Beneficial bacteria can elicit a suite of microbial and host responses to enhance the resistance to bleaching disease.
Assuntos
Microbiota , Alga Marinha , Vibrio alginolyticus/genética , Bactérias , AquiculturaRESUMO
Pathogenic bacteria can trigger active defense responses in higher plants, leading to hypersensitive programmed cell death (PCD) to against those bacteria. However, related research on seaweeds is very limited. Pseudoalteromonas piscicida X-8 (PpX-8) has been identified as the pathogen that causes bleaching disease in commercially farmed Saccharina japonica. In this study, using an inoculation assay and microscopic observations, we found that the proportion of bleaching tissue pieces inoculated with PpX-8 extracellular compounds was significantly higher (p < 0.05) than that inoculated with heated extracellular compounds, indicating that the virulence factors of PpX-8 exist in extracellular compounds and they are heat-sensitive. Using TEM, we observed typical morphological characteristics of PCD after inoculation with extracellular compounds, including chloroplast shrinkage, cytoplasmic vacuolation, and intact mitochondrial structures. Moreover, we detected biochemical characteristics of PCD, such as 3'-OH ends resulting from DNA cleavage and caspase-3-like enzymatic activity, using a TUNEL assay and fluorescence staining. Therefore, PpX-8 extracellular compounds can induce PCD, thus triggering active defense responses in S. japonica. These results indicate that seaweeds and higher plants are conservative in their active defense responses against pathogenic bacteria. The results of this study lay the foundation for further investigation of the virulence mechanisms of PpX-8.