Search Results (2)

Ice avalanche (IA) has a strong concealment and sudden characteristics, which can cause severe disasters. The early identification of IA hidden danger is of great value for disaster prevention and mitigation. However, it is very difficult, and there is poor efficiency in identifying it by site investigation or manual remote sensing. So, an artificial intelligence method for the identification of IA hidden dangers using a deep learning model has been proposed, with the glacier area of the Yarlung Tsangpo River Gorge in Nyingchi selected for identification and validation. First, through engineering geological investigations, three key identification indices for IA hidden dangers are established, glacier source, slope angle, and cracks. Sentinel-2A satellite data, Google Earth, and ArcGIS are used to extract these indices and construct a feature dataset for the study and validation area. Next, key performance metrics, such as training accuracy, validation accuracy, test accuracy, and loss rates, are compared to assess the performance of the ResNet50 (Residual Neural Network 50) and VGG16 (Visual Geometry Group 16) models. The VGG16 model (96.09% training accuracy) is selected and optimized, using Early Stopping (ES) to prevent overfitting and L2 regularization techniques (L2) to add weight penalties, which constrained model complexity and enhanced simplicity and generalization, ultimately developing the ES-L2-VGG16 (Early Stopping—L2 Norm Regularization Techniques—Visual Geometry Group 16) model (98.61% training accuracy). Lastly, during the validation phase, the model is applied to the Yarlung Tsangpo River Gorge glacier area on the Tibetan Plateau (TP), identifying a total of 100 IA hidden danger areas, with average slopes ranging between 34° and 48°. The ES-L2-VGG16 model achieves an accuracy of 96% in identifying these hidden danger areas, ensuring the precise identification of IA dangers. This study offers a new intelligent technical method for identifying IA hidden danger, with clear advantages and promising application prospects. Full article

We consider the problem of computing a set of meaningful alternative origin-to-destination routes, in real-world road network instances whose arcs are accompanied by travel-time functions rather than fixed costs. In this time-dependent alternative route scenario, we present a novel query algorithm, called Time-Dependent Alternative Graph (TDAG), that exploits the outcome of a time-consuming preprocessing phase to create a manageable amount of travel-time metadata, in order to provide answers for arbitrary alternative-routes queries, in only a few milliseconds for continental-size instances. The resulting set of alternative routes is aggregated in the form of a time-dependent alternative graph, which is characterized by the minimum route overlap, small stretch factor, small size, and low complexity. To our knowledge, this is the first work that deals with the time-dependent setting in the framework of alternative routes. The preprocessed metadata prescribe the minimum travel-time informations between a small set of “landmark” nodes and all other nodes in the graph. The TDAG query algorithm carries out the work in two distinct phases: initially, a collection phase constructs candidate alternative routes; consequently, a pruning phase cautiously discards uninteresting or low-quality routes from the candidate set. Our experimental evaluation on real-world, time-dependent road networks demonstrates that TDAG performed much better (by one or two orders of magnitude) than the existing baseline approaches. Full article