Search

• A new decision tree with preference is proposed to build a feature-based detector for small target detection in sea clutter. • Novel splitting and pruning processes are developed to meet the unbalanced demands of error probabilities and samples in radar target detection. • The proposed detector breaks the dimension limitation and has good performance and generalizability. It is difficult for maritime high-resolution radars to realize the small target detection in sea clutter, due to weak target returns and complicated clutter characteristics. Cooperation of multiple features is a recognized way to distinguish target returns from clutter. Therefore, it becomes crucial to build a detector, a special classifier, with unbalanced training samples, i.e., ergodic clutter versus non-ergodic target samples. In this paper, a preferential decision tree (pre-decision tree) with the oblique stopping criterion is proposed, where a preferential Gini index (pre-Gini index) is defined to replace the Gini index and considers rigorous false alarm rates and tolerable missed probabilities for radar target detection. Then, an improved pruning is added to the pre-decision tree to generate a dominant clutter tree, which can accurately control the false alarm rate. The two-step decision is based on the anomaly detection framework and solves the unbalance of the training samples. The proposed method can work in the high-dimensional space directly, and its decision only involves linear operations. The experimental results on the recognized IPIX and CSIR databases illustrate that the proposed method performs well among the available feature-based detectors. [ABSTRACT FROM AUTHOR]

Seasonal snow-derived water is a critical component of the water supply in the mountains and downstream regions, and the accurate characterization of available water in the form of snow-water-equivalent (SWE), peak SWE, and snowmelt onset are essential inputs for water management efforts. Arising from recent advancements in artificial intelligence (AI) and machine learning (ML), we introduce a large-scale ML SWE model leveraging publicly available data sources and open-source software. The model demonstrates the application of a limited feature space in a relatively simple ML architecture without the need for process-based formulations to effectively estimate spatially continuous SWE at a daily temporal resolution. Beginning with in situ SWE measurements (i.e., SNOTEL), lidar-derived terrain features, and temporal variables, we employ localized feature engineering and optimization via gradient-boosting decision trees to identify regionally unique drivers of snowpack dynamics and use the optimal features to train regionally independent artificial neural networks to estimate regional SWE at a 1 km spatial resolution. The model results yield respectable skill in reconstructed 1 km gridded SWE magnitudes in a hindcast simulation of the 2019 water year that is independent of the training and testing data. Comparing model estimates to over 6200 observations, the model demonstrates a weighted RMSE of 15.4 cm, Kling-Gupta Efficiency metric of 0.86, and a percent bias of 0.71% across 23 snow-influenced regions in the western U.S. The model simulation produces peak SWE estimates within 10 cm for twenty of the twenty-three regions, demonstrating capability in effectively capturing regional snow accumulation processes. The demonstration of low-error ML workflows capable of providing near-real-time, spatially continuous SWE estimates at a high spatial resolution provides proof-of-concept and a foundation to effectively update snow state variables that drive water supply forecasts in snow-dominated regions. • A new Machine Learning-based SWE estimation model for the Western United States. • Utilization of open-source tools and up-to-date in-situ observations to estimate SWE. • Reconstructed SWE magnitudes exhibit respectable skill in a hindcast simulation. • Output may serve as a supplemental state-of-snowpack input to hydrologic models. [ABSTRACT FROM AUTHOR]