Results 61 to 70 of about 3,671 (159)
We present an AI‐assisted ground‐penetrating radar framework to noninvasively detect and quantify smoldering wildfire‐induced subsurface cavities in peatlands. A ResNet model accurately predicts cavity geometry from GPR images, enabling improved assessment of underground fire damage and hazards.
Zifan Zhang +5 more
wiley +1 more source
WAH‐i: Optimising microphone array geometry for customised localisation accuracy
Abstract Accurate spatial localisation of free‐flying echolocating bats is foundational for resolving fine‐scale flight and echolocation behaviour, prey interception and spatial decision‐making in natural environments. Acoustic localisation using microphone arrays is widely employed for this purpose, yet array geometries in field studies are typically ...
Ravi Umadi
wiley +1 more source
Abstract Observations of three‐dimensional (3D) coarse root system architecture remain limited because most measurement approaches require excavation. Here, we combined dense single‐ and multi‐channel 500 MHz ground‐penetrating radar (GPR) surveys with a flow‐guided connectivity framework to non‐destructively characterise the coarse root architecture ...
Brady Flinchum +2 more
wiley +1 more source
UNIQAT: An open‐source toolkit for reproducible image quality assessment in marine surveys
Abstract Modern marine surveys generate tens of thousands of underwater images per expedition, yet many are unusable due to image quality deterioration. Manual quality screening is labour‐intensive, subjective and inconsistent across observers and teams, hindering reproducible science, creating large backlogs of unprocessed imagery, and limiting real ...
Alzayat Saleh, Arjun Chennu
wiley +1 more source
Four Questions for Ethnographers Under Trumpian Authoritarianism
ABSTRACT This article examines evolving challenges and ethical dilemmas facing ethnographers working under intensifying authoritarianism in the United States. Because very little of what the United States is experiencing is new, my work here is primarily curatorial in nature, drawing upon the lessons offered by scholars across disciplines in the United
Elizabeth Hanna Rubio
wiley +1 more source
Multi‐Spectral Gaussian Splatting with Neural Color Representation
Abstract 3D Gaussian Splatting (3DGS) [KKLD23] has transformed novel‐view synthesis from RGB images, yet remains restricted to the visible spectrum. Many applications, including agricultural monitoring, rely on multi‐spectral imaging, where spectral camera alignment and scalability pose major challenges.
Lukas Meyer +5 more
wiley +1 more source
Class Angular Distortion Index for Dimensionality Reduction
Abstract Dimensionality reduction (DR) techniques are often characterized by whether they preserve global, high‐level structures in the data or local, neighborhood structures. This distinction matters in visualization: global methods can obscure clusters while local methods can over‐emphasize them.
Kaviru Gunaratne +2 more
wiley +1 more source
Fairness‐aware insurance pricing: A multi‐objective optimization approach
Abstract Machine‐learning models can provide accurate predictions in insurance pricing, but can also increase disparities between protected groups. Existing fairness‐aware pricing approaches typically target one fairness notion at a time, making it difficult to compare trade‐offs between predictive accuracy, group fairness, individual fairness, and ...
Tim J. Boonen, Xinyue Fan, Zixiao Quan
wiley +1 more source
Unmarked Emotional States and the Affective Anchoring of Continuity
ABSTRACT Narratives around emotions often foreground remarkable episodes that interrupt situations, producing a “rollercoaster” image of emotional life that leaves its stability underdescribed. To analyze the emotional dimension of social continuity, this article theorizes unmarked emotional states (UES): culturally default, interactionally unobtrusive
Lorenzo Sabetta
wiley +1 more source
Evaluating Large‐Language Models in Bioinformatics Applications
This work systematically evaluates advanced large language models across six representative bioinformatics tasks, from drug–drug interaction prediction and molecular optimization to single‐cell annotation and computational biology problem solving. The benchmark quantifies both their promise and limitations, offering practical guidance for prompt design,
Hengchuang Yin +9 more
wiley +1 more source

