Search

As a variant task of time-series segmentation, trajectory segmentation is a key task in the applications of transportation pattern recognition and traffic analysis. However, segmenting trajectory is faced with challenges of implicit patterns and sparse results. Although deep neural networks have tremendous advantages in terms of high-level feature learning performance, deploying as a blackbox seriously limits the real-world applications. Providing explainable segmentations has significance for result evaluation and decision making. Thus, in this article, we address trajectory segmentation by proposing a Bayesian Encoder-Decoder Network (BED-Net) to provide accurate detection with explainability and references for the following active-learning procedures. BED-Net consists of a segmentation module based on Monte Carlo dropout and an explanation module based on uncertainty learning that provides results evaluation and visualization. Experimental results on both benchmark and real-world datasets indicate that BED-Net outperforms the rival methods and offers excellent explainability in the applications of trajectory segmentation. [ABSTRACT FROM AUTHOR]

Irrigation mode is an important factor in regulating CH 4 and N 2 O emissions from croplands. However, there are no studies on the effects of irrigation mode practiced during the rice season on the annual integrative global warming potential (GWP) of CH 4 and N 2 O in the rice–winter wheat rotation systems. Thus, a field experiment was designed to study the effects of controlled irrigation (CI) during the rice season on the annual integrative GWP of CH 4 and N 2 O emissions from the rice–winter wheat rotation systems, with traditional irrigation (TI) as the control. A notable trade-off relationship between CH 4 and N 2 O emissions was observed in the rice–wheat rotation systems under both CI and TI of rice. CI during the rice season had obvious subsequent effects on CH 4 and N 2 O emissions from the following winter wheat season. Over the whole annual cycle, CI significantly reduced the cumulative CH 4 emission (13.27 kg ha −1 ) by 80.6% than that from the TI fields ( p < 0.001), but did not cause a significant difference on the cumulative N 2 O emission ( p > 0.05). The integrative GWP of CH 4 and N 2 O on a 100-year horizon for the CI rotation systems was 2720.35 kg CO 2 ha −1 , which was 41.1% lower than that for the TI fields ( p < 0.05). Moreover, the difference of rice and wheat yields between the CI and TI fields was not significant ( p > 0.05). This is the first study to investigate the effects of CI during the rice season on the annual integrative GWP of CH 4 and N 2 O in a rice–winter wheat rotation system. Our results suggest that CI can significantly mitigate the annual integrative greenhouse effect caused by CH 4 and N 2 O from the rice–winter wheat rotation systems, while ensuring the crop yields. [ABSTRACT FROM AUTHOR]