Investigation of Wildfires at Forested Landscapes: A Novel Contribution to Nonparametric Density Mapping at Regional Scale

Kuter, Nazan; Kuter, Semih

doi:10.15666/aeer/1604_47014716

Cited by 4 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…MARS-based KDE approach generates a single smoothing parameter instead of several values over a certain fixed interval. 34 However, atmospheric correction models are constructed using conic multivariate adaptive regression splines to overcome the drawbacks of MARS. 35 Multiresponse MARS approach is adopted for snow mapping on MODIS images.…”

Section: Related Workmentioning

confidence: 99%

Mixed pixel removal in north Tamil Nadu region for accurate area measurement

Mani¹,

Nagalingam²,

Aruldoss³

et al. 2021

Computational Intelligence

View full text Add to dashboard Cite

In remote sensing, vegetation and water areas delineation from satellite image plays a vital role for urban and rural planning. Delineation of vegetation and water area is a challenging task due to mixed pixels and geometric distortion over the boundary region. Geometric distortion arises due to change in velocity and speed of satellite during image acquisition, and mixed pixels arises due to different surfaces in a particular area. Traditional methods apply classifier algorithms such as support vector machine, neural network, and fuzzy for vegetation and water area delineation. The traditional methods require more training dataset and consume more interpretation time for delineation. In this article, we propose transverse dyadic wavelet transform (TDyWT) to delineate vegetation and water area from Landsat 8 images. The TDyWT method enhances the boundary and curvature area of satellite image for accurate delineation. From the experimental results, the proposed TDyWT approach delineates the area of subclass for vegetation and water areas with 95% of accuracy with respect to the ground truth.

show abstract

Section: Related Workmentioning

confidence: 99%

Mixed pixel removal in north Tamil Nadu region for accurate area measurement

Mani¹,

Nagalingam²,

Aruldoss³

et al. 2021

Computational Intelligence

View full text Add to dashboard Cite

show abstract

“…There is much work in the literature already where statistical methods have been used to create a spatial map of some property of interest. In disciplines such as environmental science and neuroscience, a nonparametric regression method, namely multivariate adaptive regression splines (MARS), has allowed researchers to create spatial maps of, for example, forest fires and neurological events [18][19][20][21] and it has been shown that MARS is able to cope with large and complex data sets [19]. Furthermore, this technique is flexible and adaptive and as a result, it can be applied to linear and non-linear problems [19].…”

Section: Introductionmentioning

confidence: 99%

“…Furthermore, this technique is flexible and adaptive and as a result, it can be applied to linear and non-linear problems [19]. There has been some success with this technique in reducing the positional uncertainties of the data points by reducing the discretiszation of the geometry [18]. However, due to the differences in the application, the discretization in [18] is still much larger than in the metrological setting central to this paper.…”

Section: Introductionmentioning

confidence: 99%

Reconstruction of refractive index maps using photogrammetry

Miller

Mullholland

Tant

et al. 2021

Inverse Problems in Science and Engineering

View full text Add to dashboard Cite

Large volume metrology is a key enabler of autonomous precision manufacturing. For component positioning, the optical-based metrology technique of photogrammetry could be used more widely if its accuracy was improved. These positional measurements are subject to uncertainties which can be greater than manufacturing tolerances. One source of uncertainty is due to thermal gradients, which cause the refraction of the light rays in large-scale industrial environments. This paper uses light-based sensor data to reconstruct a heterogeneous spatial map of the refractive index in air. We use this reconstructed refractive index map to discount the refractive effects and thereby reduce the uncertainty of this positioning problem. This new inverse problem employs Voronoi tessellations to spatially parameterize the refractive index map, the Fast Marching Method to solve the forward problem of calculating the light rays through this medium, and a Bayesian approach in the inversion. Using simulated data, this methodology leads to positioning improvements of up to 37%.

show abstract

Explaining Landscape Levels and Drivers of Chinese Moso Bamboo Forests Based on the Plus Model

Wei

et al. 2023

Forests

View full text Add to dashboard Cite

China is the richest country in the world in terms of bamboo forest resources, with moso bamboo as the dominated landscape distribution. Analysis of its spatial distribution, landscape change, and its drivers is crucial for forest ecosystem management and sustainable development. However, investigations on the effects of multiple geographical and environmental factors on changes in the landscape of moso bamboo forests are still limited. In this study, Chinese moso bamboo forests in 2010, 2015 and 2020 were selected as the study objects, and 19 provinces (data for Hong Kong, Macao, and Taiwan are unavailable), where Chinese moso bamboo forests were actually distributed, were taken as the study areas. This paper aims to determine the spatial distribution and landscape level of moso bamboo forests in China, as well as to conduct a preliminary study on the natural and socioeconomic factors of landscape change within moso bamboo forests and their buffer zones through density analysis, landscape fragmentation analysis, and patch-generating land use simulation model. The analysis using ArcGIS kernel density analysis revealed significant variability in the spatial distribution of moso bamboo forests in China, expanding in both the north and southwest directions. China’s moso bamboo forests expanded fast between 2010 and 2020, with the landscape becoming more fragmented, landscape fragmentation increasing, aggregation diminishing, and overall landscape quality declining. Climate has the greatest influence on the shifting landscape distribution of moso bamboo forests, followed by locational factors and soil and terrain, and socioeconomic factors such as location, population density, and GDP also impact the shifting distribution and landscape of the moso bamboo forest.

show abstract

Investigation of Wildfires at Forested Landscapes: A Novel Contribution to Nonparametric Density Mapping at Regional Scale

Cited by 4 publications

References 0 publications

Mixed pixel removal in north Tamil Nadu region for accurate area measurement

Mixed pixel removal in north Tamil Nadu region for accurate area measurement

Reconstruction of refractive index maps using photogrammetry

Explaining Landscape Levels and Drivers of Chinese Moso Bamboo Forests Based on the Plus Model

Contact Info

Product

Resources

About