Search

The health of the trees in the forest affects the ecological environment, so timely detection of Standing Dead Trees (SDTs) plays an important role in forest management. However, due to the large spatial scope of forests, it is difficult to find SDTs through conventional approaches such as field inventories. In recent years, the development of deep learning and Unmanned Aerial Vehicle (UAV) has provided technical support for low-cost real-time monitoring of SDTs, but the inability to fully utilize global features and the difficulty of small-scale SDTs detection have brought challenges to the detection of SDTs in visible light images. Therefore, this paper proposes a multi-scale attention mechanism detection method for identifying SDTs in UAV RGB images. This method takes Faster-RCNN as the basic framework and uses Swin-Transformer as the backbone network for feature extraction, which can effectively obtain global information. Then, features of different scales are extracted through the feature pyramid structure and feature balance enhancement module. Finally, dynamic training is used to improve the quality of the model. The experimental results show that the algorithm proposed in this paper can effectively identify the SDTs in the visible light image of the UAV with an accuracy of 95.9%. This method of SDTs identification can not only improve the efficiency of SDTs exploration, but also help relevant departments to explore other forest species in the future. [ABSTRACT FROM AUTHOR]

Influenza is active during the winter and spring in the city of Beijing, which has a typical temperate climate with four clear distinct seasons. The clinical and laboratory surveillance data for influenza have been used to construct critical indicators for influenza activities in the community, and previous studies have reported varying degrees of association between laboratory-confirmed influenza specimens and outpatient consultation rates of influenza-like illness in subtropical cities. However, few studies have reported on this issue for cities in temperate regions, especially in developing countries. Furthermore, the mechanism behind age-specific seasonal epidemics remains unresolved, although it has been widely discussed. We utilized a wavelet analysis method to monitor the coherence of weekly percentage of laboratory-confirmed influenza specimens with the weekly outpatient consultation rates of influenza-like illness in Beijing, China. We first examined the seasonal pattern of laboratory-confirmed cases of influenza A (subtyped into seasonal A(H1N1) and A(H3N2) and pandemic virus A(H1N1) pdm09) and influenza B separately within the period from 2008–2015; then, we detected the coherence of clinical and laboratory surveillance data in this district, specially examining weekly time series of age-specific epidemics of influenza-like illnesses in the whole study period for three age categories (age 0–5, 5–15 and 25–60). We found that influenza A and B were both active in winter but were not always seasonally synchronous in Beijing. Synchronization between age ranges was found in most epidemic peaks from 2008–2015. Our findings suggested that peaks of influenza-like illness in individuals aged 0–5 and 5–15 years consistently appeared ahead of those of adults, implying the possibility that schoolchildren may lead epidemic fluctuations. [ABSTRACT FROM AUTHOR]

Student Dropout Prediction (SDP) is pivotal in mitigating withdrawals in Massive Open Online Courses. Previous studies generally modeled the SDP problem as a binary classification task, providing a single prediction outcome. Accordingly, some attempts introduce survival analysis methods to achieve continuous and consistent predictions over time. However, the volatility and sparsity of data always weaken the models' performance. Prevailing solutions rely heavily on data pre-processing independent of predictive models, which are labor-intensive and may contaminate authentic data. This paper proposes a Survival Analysis based Volatility and Sparsity Modeling Network (SAVSNet) to address these issues in an end-to-end deep learning framework. Specifically, SAVSNet smooths the volatile time series by convolution network while preserving the original data information using Long-Short Term Memory Network (LSTM). Furthermore, we propose a Time-Missing-Aware LSTM unit to mitigate the impact of data sparsity by integrating informative missingness patterns into the model. A survival analysis loss function is adopted for parameter estimation, and the model outputs monotonically decreasing survival probabilities. In the experiments, we compare the proposed method with state-of-the-art methods in two real-world MOOC datasets, and the experiment results show the effectiveness of our proposed model. [ABSTRACT FROM AUTHOR]