Segmentation and terrain modelling of extra‐terrestrial chasmata

Miliaresis, George; Kokkas, Nikolaos

doi:10.1080/14498596.2004.9635024

Cited by 7 publications

(5 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Slope aspect mapping is classified as a continuous terrain segmentation scheme (Miliaresis and Kokkas, 2004). In the current approach, aspect was computed for every DEM grid point (Figure 3C).…”

Section: Aspect Modelingmentioning

confidence: 99%

Tectonic Geomorphology and Volcano-Tectonic Interaction in the Eastern Boundary of the Southern Cascades (Hat Creek Graben Region), California, USA

Paguican

Bursik

2016

Front. Earth Sci.

View full text Add to dashboard Cite

The eastern boundary of the Southern Cascades (Hat Creek Graben region), California, USA, is an extensively faulted volcanic corridor between the Cascade Range and Modoc Plateau. The morphology of the region is a result of plate motions associated with different tectonic provinces, faulting, and recurring volcanic activity, making it an ideal place to study the interrelationship between tectonics, volcanoes, and geomorphology. We use the morphometry and spatial distribution of volcanoes and their interaction with regional structures to understand how long term regional deformation can affect volcano evolution. A database of volcanic centers and structures was created from interpretations of digital elevation models. Volcanic centers were classified by morphological type into cones, sub-cones, shields and massifs. A second classification by height separated the larger and smaller edifices, and revealed an evolutionary trend. Poisson Nearest Neighbor analysis showed that bigger volcanoes are spatially dispersed while smaller ones are clustered. Using volcano centroid locations, about 90 lineaments consisting of at least three centers within 6 km of one another were found, revealing that preferential north-northwest directed pathways control the transport of magma from the source to the surface, consistent with the strikes of the major fault systems. Most of the volcano crater and collapse scar openings are perpendicular to the north northwest-directed maximum horizontal stress, expected for extensional environments with dominant normal faulting. Early in the history of a volcano or volcano cluster, melt propagates to the surface using the easiest and most efficient pathway, mostly controlled by the pre-existing normal faults and near-surface stress fields, as indicated by the pervasive vent alignments. Volcano growth continues to be dependent on the regional structures as indicated by the opening directions, suggesting structural control on the growth of the volcanic edifices. The results present a particularly well-defined case in which extension of a volcanic region is accommodated mostly by faulting, and only partly by intrusion to form volcanoes. This is attributed to a low magma supply rate.

show abstract

Section: Aspect Modelingmentioning

confidence: 99%

Tectonic Geomorphology and Volcano-Tectonic Interaction in the Eastern Boundary of the Southern Cascades (Hat Creek Graben Region), California, USA

Paguican

Bursik

2016

Front. Earth Sci.

View full text Add to dashboard Cite

show abstract

“…The basic approach to terrain segmentation is to define a basic set of terrain classes and apply a standard machine learning algorithm such as Support Vector Machines (SVM) [16] or Convolutional Neural Networks (CNN) [15]. The segmentation results can be further refined by applying Fast Fourier Transforms (FFT) [17] or connected component labelling [18]. TU-Net and TDeepLab have been proposed to allow terrain segmentation to be more robust to illumination changes [6], but they rely on data fusion with thermal images, which is not always available.…”

Section: Terrain Segmentationmentioning

confidence: 99%

Mars Terrain Segmentation with Less Labels

Goh¹,

Chen²,

Wilson³

2022

Preprint

View full text Add to dashboard Cite

Planetary rover systems need to perform terrain segmentation to identify drivable areas as well as identify specific types of soil for sample collection. The latest Martian terrain segmentation methods rely on supervised learning which is very data hungry and difficult to train where only a small number of labeled samples are available. Moreover, the semantic classes are defined differently for different applications (e.g., rover traversal vs. geological) and as a result the network has to be trained from scratch each time, which is an inefficient use of resources. This research proposes a semi-supervised learning framework for Mars terrain segmentation where a deep segmentation network trained in an unsupervised manner on unlabeled images is transferred to the task of terrain segmentation trained on few labeled images. The network incorporates a backbone module which is trained using a contrastive loss function and an output atrous convolution module which is trained using a pixel-wise cross-entropy loss function. Evaluation results using the metric of segmentation accuracy show that the proposed method with contrastive pretraining outperforms plain supervised learning by 2%-10%. Moreover, the proposed model is able to achieve a segmentation accuracy of 91.1% using only 161 training images (1% of the original dataset) compared to 81.9% with plain supervised learning.

show abstract

“…Aspect regions are a characteristic of a continuous terrain segmentation scheme (Miliaresis and Kokkas, 2004). In this approach, aspect is computed for every DEM point (Figure 4) and then standardized to eight directions (N, NE, E, SE, S, SW, W, and NW) defined in a raster image (Table 1; Figure 5d).…”

Section: Terrain Segmentationmentioning

confidence: 99%

“…Terrain modelling involves the use of segmentation techniques to classify landscapes and to measure size, shape, and mutual relations between recognized landforms (Pike, 1993;. Terrain segmentation, performed in either a discontinuous or continuous framework, allows us to identify and map landforms with similar or contrasting ranges of characteristics (Iwahashi and Pike, 2007), and the parametric representation of the obtained terrain classification, based on its three-dimensional (3D) spatial pattern, provides useful information about geomorphological features (Miliaresis and Kokkas, 2004;Miliaresis, 2006). In addition, a continuous terrain partition scheme based on aspect has been applied in region classification based on geomorphometric criteria (Dymond et al, 1995).…”

Section: Introductionmentioning

confidence: 99%

Full Issue in PDF / Numéro complet enform PDF

2009

Canadian Journal of Remote Sensing

View full text Add to dashboard Cite

Segmentation and terrain modelling of extra‐terrestrial chasmata

Cited by 7 publications

References 14 publications

Tectonic Geomorphology and Volcano-Tectonic Interaction in the Eastern Boundary of the Southern Cascades (Hat Creek Graben Region), California, USA

Tectonic Geomorphology and Volcano-Tectonic Interaction in the Eastern Boundary of the Southern Cascades (Hat Creek Graben Region), California, USA

Mars Terrain Segmentation with Less Labels

Full Issue in PDF / Numéro complet enform PDF

Contact Info

Product

Resources

About