Search

This paper addresses the challenge of training deep learning models for fire smoke scene detection from multi-sensor, multi-spectral satellite imagery, where spectral bands vary and training data is scarce, especially for new sensors. We introduce a novel cross-sensor transfer learning approach enhanced by spectral pattern extraction, notably aided by a deep learning (DL) module called Input Amplification (IA). Our contribution includes the creation of two datasets, Landsat_smk2770 and Sentinel2_smk, with varying spectral bands. We employ a model named IA_VIB_SD which incorporates IA and is designed specifically for fire smoke scene detection. We investigated the transferability of the Landsat_smk2770-trained IA_VIB_SD model to the Sentinel2_smk dataset using various transfer learning techniques, and compared the performance of the transferred model with the benchmark IA_VIB_SD model exclusively trained on pure Sentinel2_smk data. Our proposed transfer learning approach resulted in a transferred model with an average accuracy 5% higher than the benchmark model. Notably, Our proposed transfer learning approach outperformed existing transfer learning methods by more than 1% in terms of the accuracy, even when trained on a mere 10% of the Sentinel2_smk dataset. This research advances fire smoke scene detection from multi-sensor, multi-spectral satellite imagery. [ABSTRACT FROM AUTHOR]

We present the GLIMPSE system that provides a framework for storage, management, accessibility and integration of 3D LiDAR data acquired from multiple platforms. We detail a point cloud retrieval approach, which provides spatially optimised access to point cloud data from the system for a particular geographic area based on user specifications. We tested our point cloud retrieval approach to facilitate the extraction of road medians from large volumes of ALS data stored in the GLIMPSE system. The integrated use of a geospatial database, the GLIMPSE system and the point cloud retrieval approach improved the efficiency of road median extraction. [ABSTRACT FROM AUTHOR]

Comprehensive spatial plans require a wide range of high quality spatial data, but planning requirements are not always considered in the increasing number of spatial data initiatives. This paper presents a framework for systematically analysing the availability and quality of spatial datasets in urban planning, and applies this to the example of the metropolitan strategy for Greater Adelaide, South Australia. The findings reveal significant gaps in important spatial data that are publicly available. This is especially concerning given the need to better understand integrated and multi-scalar planning policies such as for mixed use developments and polycentric urban structures. [ABSTRACT FROM AUTHOR]

Purpose: To explore the experiences and perspectives of the healthcare professionals who were trained to and delivered "Minimise Fatigue, Maximise Life" (MFML), a patient-centered group-based fatigue self-management program for persons with multiple sclerosis. Methods: A qualitative descriptive study with semi-structured individual interviews at two time points. Data were analyzed for themes. Six healthcare professional facilitators who were trained to and delivered "Minimise Fatigue, Maximise Life" participated in a first interview, and five in a second. Participants were all female, aged between 23 and 66 years old and either occupational therapists or physiotherapists. Results: Two themes were evident in the data. The first, "Reciprocity," showed how the healthcare professionals were trained to deliver MFML, then reciprocated in the program delivery as active participants, which then provided feelings of personal reward and expansion of their usual practice. The second, "Enhancements," encompassed suggested directions for future training and deliveries of the program. Conclusion: This study suggests that multidimensional patient centered interventions also benefit the healthcare professionals who provide them because it expands their practice. Healthcare professionals who recognize the benefits of innovative and patient-centered interventions, supports both the patients with whom they work, and adds value to the health services they provide. Healthcare professionals who undergo training to facilitate delivery of self-management programs, which are based in an empowerment model, report an enhancement or expansion of their traditional practice. An empowerment-based program delivered in a group situation encourages and facilitates people to draw on their own and peers' knowledge and expertise to problem solve for self-management. Healthcare professional education should facilitate the healthcare professional's learning, and ability and willingness to acknowledge the richness in knowledge and expertise held by their patients. [ABSTRACT FROM AUTHOR]

In this article, we derive the inside–outside duality for two time-harmonic, elastic scattering problems. First, we consider a rigid scattering object inside an isotropic, homogeneous background medium and second, we consider a penetrable, inhomogeneous scattering object inside this background medium. For the first scattering problem, we make use of the particular behavior or certain eigenvalues of the corresponding far field operator to characterize interior Dirichlet eigenvalues of the negative Navier operator. Then we adapt this technique to determine interior transmission eigenvalues that correspond to the second scattering problem. [ABSTRACT FROM AUTHOR]

This article describes a new automatic quadtree-/octree-based and scale-dependent generalization algorithm for point selection. The benefit toward existing point selection methods is that it preserves global as well as local characteristics of the spatial point distribution and of the spatial point density. It can be applied not only to points in 2D space but also to points in 3D space. In this article, an evaluation of the new point selection method is also provided. [ABSTRACT FROM AUTHOR]

The purpose of this article is to investigate the density calculation and representation of spatially and temporally highly dynamic point data sets. We suggest an approach to explore point patterns that have a temporal dimension and therefore introduce an incremental development of the traditional kernel density estimation processes. Based on a movement vector assigned to each moving point, we apply a directed (or tilted) kernel to calculate and visualize the density pattern. The resulting density map recognizes the dynamic behavior of the underlying data points. By applying a shade effect or contour lines, the areas with densely distributed moving points are characterized by directed ‘ripples’ or ‘waves’. This assists the visual analysis and prediction of ‘movement trends’ based on the dynamic points. In our case study, we apply this method to airplane movement data limited to two points in time and we thereby visualize the results for the study area over Germany. [ABSTRACT FROM AUTHOR]