Given N geo-located point instances (e.g., crime or disease cases) in a spatial domain, we aim to detect sub-regions (i.e., hotspots) that have a higher probability density of generating such instances than the others. Hotspot detection has been widely used in a variety of important urban applications, including public safety, public health, urban planning, equity, etc. The problem is challenging because its societal applications often have low-tolerance for false positives, and require significance testing which is computationally intensive. In related work, the spatial scan statistic introduced a likelihood ratio based framework for hotspot evaluation and significance testing. However, it fails to consider the effect of spatial nondeterminism, causing many missing detections. Our previous work introduced a nondeterministic normalization based scan statistic to mitigate this issue. However, its robustness against false positives is not stably controlled. To address these limitations, we propose a unified framework which can improve the completeness of results without incurring more false positives. We also propose a reduction algorithm to improve the computational efficiency. Experiment results confirm that the unified framework can greatly improve the recall of hotspot detection without increasing the number of false positives, and the reduction algorithm can greatly reduce execution time.
more »
« less
This content will become publicly available on April 26, 2025
Automatic identification and quantification of volcanic hotspots in Alaska using HotLINK: the hotspot learning and identification network
An increase in volcanic thermal emissions can indicate subsurface and surface processes that precede, or coincide with, volcanic eruptions. Space-borne infrared sensors can detect hotspots—defined here as localized volcanic thermal emissions—in near-real-time. However, automatic hotspot detection systems are needed to efficiently analyze the large quantities of data produced. While hotspots have been automatically detected for over 20 years with simple thresholding algorithms, new computer vision technologies, such as convolutional neural networks (CNNs), can enable improved detection capabilities. Here we introduce HotLINK: the Hotspot Learning and Identification Network, a CNN trained to detect hotspots with a dataset of −3,800 satellite-based, Visible Infrared Imaging Radiometer Suite (VIIRS) images from Mount Veniaminof and Mount Cleveland volcanoes, Alaska. We find that our model achieves an accuracy of 96% (F1-score 0.92) when evaluated on −1,700 unseen images from the same volcanoes, and 95% (F1-score 0.67) when evaluated on −3,000 images from six additional Alaska volcanoes (Augustine Volcano, Bogoslof Island, Okmok Caldera, Pavlof Volcano, Redoubt Volcano, Shishaldin Volcano). In comparison with an existing threshold-based hotspot detection algorithm, MIROVA (Coppola et al., Geological Society, London, Special Publications, 2016, 426, 181–205), our model detects 22% more hotspots and produces 12% fewer false positives. Additional testing on −700 labeled Moderate Resolution Imaging Spectroradiometer (MODIS) images from Mount Veniaminof demonstrates that our model is applicable to this sensor’s data as well, achieving an accuracy of 98% (F1-score 0.95). We apply HotLINK to 10 years of VIIRS data and 22 years of MODIS data for the eight aforementioned Alaska volcanoes and calculate the radiative power of detected hotspots. From these time series we find that HotLINK accurately characterizes background and eruptive periods, similar to MIROVA, but also detects more subtle warming signals, potentially related to volcanic unrest. We identify three advantages to our model over its predecessors: 1) the ability to detect more subtle volcanic hotspots and produce fewer false positives, especially in daytime images; 2) probabilistic predictions provide a measure of detection confidence; and 3) its transferability, i.e., the successful application to multiple sensors and multiple volcanoes without the need for threshold tuning, suggesting the potential for global application.
more » « less- Award ID(s):
- 1855126
- NSF-PAR ID:
- 10518543
- Publisher / Repository:
- Frontiers in Earth Science
- Date Published:
- Journal Name:
- Frontiers in Earth Science
- Volume:
- 12
- ISSN:
- 2296-6463
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)The Cook-Austral volcanic lineament extends from Macdonald Seamount (east) to Aitutaki Island (west) in the South Pacific Ocean and consists of hotspot-related volcanic islands, seamounts, and atolls. The Cook-Austral volcanic lineament has been characterized as multiple overlapping, age-progressive hotspot tracks generated by at least two mantle plumes, including the Arago and Macdonald plumes, which have fed volcano construction for ~20 m.y. The Arago and Macdonald hotspot tracks are argued to have been active for at least 70 m.y. and to extend northwest of the Cook-Austral volcanic lineament into the Cretaceous-aged Tuvalu-Gilbert and Tokelau Island chains, respectively. Large gaps in sampling exist along the predicted hotspot tracks, complicating efforts seeking to show that the Arago and Macdonald hotspots have been continuous, long-lived sources of hotspot volcanism back into the Cretaceous. We present new major- and trace-element concentrations and radiogenic isotopes for three seamounts (Moki, Malulu, Dino) and one atoll (Rose), and new clinopyroxene 40Ar/39Ar ages for Rose (24.81 ± 1.02 Ma) and Moki (44.53 ± 10.05 Ma). All volcanoes are located in the poorly sampled region between the younger Cook-Austral and the older, Cretaceous portions of the Arago and Macdonald hotspot tracks. Absolute plate motion modeling indicates that the Rose and Moki volcanoes lie on or near the reconstructed traces of the Arago and Macdonald hotspots, respectively, and the 40Ar/39Ar ages for Rose and Moki align with the predicted age progression for the Arago (Rose) and Macdonald (Moki) hotspots, thereby linking the younger Cook-Austral and older Cretaceous portions of the long-lived (>70 m.y.) Arago and Macdonald hotspot tracks.more » « less
-
more » « less
Abstract Since the 1919 foundation of the International Association of Volcanology and Chemistry of the Earth’s Interior (IAVCEI), the fields of volcano seismology and acoustics have seen dramatic advances in instrumentation and techniques, and have undergone paradigm shifts in the understanding of volcanic seismo-acoustic source processes and internal volcanic structure. Some early twentieth-century volcanological studies gave equal emphasis to barograph (infrasound and acoustic-gravity wave) and seismograph observations, but volcano seismology rapidly outpaced volcano acoustics and became the standard geophysical volcano-monitoring tool. Permanent seismic networks were established on volcanoes (for example) in Japan, the Philippines, Russia, and Hawai‘i by the 1950s, and in Alaska by the 1970s. Large eruptions with societal consequences generally catalyzed the implementation of new seismic instrumentation and led to operationalization of research methodologies. Seismic data now form the backbone of most local ground-based volcano monitoring networks worldwide and play a critical role in understanding how volcanoes work. The computer revolution enabled increasingly sophisticated data processing and source modeling, and facilitated the transition to continuous digital waveform recording by about the 1990s. In the 1970s and 1980s, quantitative models emerged for long-period (LP) event and tremor sources in fluid-driven cracks and conduits. Beginning in the 1970s, early models for volcano-tectonic (VT) earthquake swarms invoking crack tip stresses expanded to involve stress transfer into the wall rocks of pressurized dikes. The first deployments of broadband seismic instrumentation and infrasound sensors on volcanoes in the 1990s led to discoveries of new signals and phenomena. Rapid advances in infrasound technology; signal processing, analysis, and inversion; and atmospheric propagation modeling have now established the role of regional (15–250 km) and remote (> 250 km) ground-based acoustic systems in volcano monitoring. Long-term records of volcano-seismic unrest through full eruptive cycles are providing insight into magma transport and eruption processes and increasingly sophisticated forecasts. Laboratory and numerical experiments are elucidating seismo-acoustic source processes in volcanic fluid systems, and are observationally constrained by increasingly dense geophysical field deployments taking advantage of low-power, compact broadband, and nodal technologies. In recent years, the fields of volcano geodesy, seismology, and acoustics (both atmospheric infrasound and ocean hydroacoustics) are increasingly merging. Despite vast progress over the past century, major questions remain regarding source processes, patterns of volcano-seismic unrest, internal volcanic structure, and the relationship between seismic unrest and volcanic processes.
-
Face touch is an unconscious human habit. Frequent touching of sensitive/mucosal facial zones (eyes, nose, and mouth) increases health risks by passing pathogens into the body and spreading diseases. Furthermore, accurate monitoring of face touch is critical for behavioral intervention. Existing monitoring systems only capture objects approaching the face, rather than detecting actual touches. As such, these systems are prone to false positives upon hand or object movement in proximity to one's face (e.g., picking up a phone). We present FaceSense, an ear-worn system capable of identifying actual touches and differentiating them between sensitive/mucosal areas from other facial areas. Following a multimodal approach, FaceSense integrates low-resolution thermal images and physiological signals. Thermal sensors sense the thermal infrared signal emitted by an approaching hand, while physiological sensors monitor impedance changes caused by skin deformation during a touch. Processed thermal and physiological signals are fed into a deep learning model (TouchNet) to detect touches and identify the facial zone of the touch. We fabricated prototypes using off-the-shelf hardware and conducted experiments with 14 participants while they perform various daily activities (e.g., drinking, talking). Results show a macro-F1-score of 83.4% for touch detection with leave-one-user-out cross-validation and a macro-F1-score of 90.1% for touch zone identification with a personalized model.more » « less
-
more » « less
Abstract In recent years, deep learning gained proliferating popularity in the cybersecurity application domain, since when being compared to traditional machine learning methods, it usually involves less human efforts, produces better results, and provides better generalizability. However, the imbalanced data issue is very common in cybersecurity, which can substantially deteriorate the performance of the deep learning models. This paper introduces a transfer learning based method to tackle the imbalanced data issue in cybersecurity using return-oriented programming payload detection as a case study. We achieved 0.0290 average false positive rate, 0.9705 average F1 score and 0.9521 average detection rate on 3 different target domain programs using 2 different source domain programs, with 0 benign training data sample in the target domain. The performance improvement compared to the baseline is a trade-off between false positive rate and detection rate. Using our approach, the total number of false positives is reduced by 23.16%, and as a trade-off, the number of detected malicious samples decreases by 0.68%.
