scholarly journals An example of SAR-derived image segmentation for landslides detection

2018 ◽  
Author(s):  
Giuseppe Esposito ◽  
Alessandro Cesare Mondini ◽  
Ivan Marchesini ◽  
Paola Reichenbach ◽  
Paola Salvati ◽  
...  
Keyword(s):  
Spatial Scales ◽  
Rapid Assessment ◽  
Ground Truth ◽  
Radar Data ◽  
Mass Movements ◽  
Ground Truth Data ◽  
Before And After ◽  
Satellite Radar ◽  
Gis Software ◽  
Derived Data

A rapid assessment of the areal extent of landslide disasters is one of the main challenges facing by the scientific community. Satellite radar data represent a powerful tool for the rapid detection of landslides over large spatial scales, even in case of persistent cloud cover. To define landslide locations, radar data need to be firstly pre-processed and then elaborated for the extraction of the required information. Segmentation represents one of the most useful procedures for identifying land cover changes induced by landslides. In this study, we present an application of the i.segment module of GRASS GIS software for segmenting radar-derived data. As study area, we selected the Tagari River valley in Papua New Guinea, where massive landslides were triggered by a M7.5 earthquake on February 25, 2018. A comparison with ground truth data revealed a suitable performance of i.segment. Particular segmentation patterns, in fact, resulted in the areas affected by landslides with respect to the external ones, or to the same areas before the earthquake. These patterns highlighted a relevant contrast of radar backscattering values recorded before and after the landslides. With our procedure, we were able to define the extension of the mass movements that occurred in the study area, just three days after the M7.5 earthquake.

scholarly journals An example of SAR-derived image segmentation for landslides detection

2018 ◽  
Author(s):  
Giuseppe Esposito ◽  
Alessandro Cesare Mondini ◽  
Ivan Marchesini ◽  
Paola Reichenbach ◽  
Paola Salvati ◽  
...  
Keyword(s):  
Spatial Scales ◽  
Rapid Assessment ◽  
Ground Truth ◽  
Radar Data ◽  
Mass Movements ◽  
Ground Truth Data ◽  
Before And After ◽  
Satellite Radar ◽  
Gis Software ◽  
Derived Data

A rapid assessment of the areal extent of landslide disasters is one of the main challenges facing by the scientific community. Satellite radar data represent a powerful tool for the rapid detection of landslides over large spatial scales, even in case of persistent cloud cover. To define landslide locations, radar data need to be firstly pre-processed and then elaborated for the extraction of the required information. Segmentation represents one of the most useful procedures for identifying land cover changes induced by landslides. In this study, we present an application of the i.segment module of GRASS GIS software for segmenting radar-derived data. As study area, we selected the Tagari River valley in Papua New Guinea, where massive landslides were triggered by a M7.5 earthquake on February 25, 2018. A comparison with ground truth data revealed a suitable performance of i.segment. Particular segmentation patterns, in fact, resulted in the areas affected by landslides with respect to the external ones, or to the same areas before the earthquake. These patterns highlighted a relevant contrast of radar backscattering values recorded before and after the landslides. With our procedure, we were able to define the extension of the mass movements that occurred in the study area, just three days after the M7.5 earthquake.

scholarly journals An example of SAR-derived image segmentation for landslides detection

2018 ◽  
Author(s):  
Giuseppe Esposito ◽  
Alessandro Cesare Mondini ◽  
Ivan Marchesini ◽  
Paola Reichenbach ◽  
Paola Salvati ◽  
...  
Keyword(s):  
Spatial Scales ◽  
Rapid Assessment ◽  
Ground Truth ◽  
Radar Data ◽  
Mass Movements ◽  
Ground Truth Data ◽  
Before And After ◽  
Satellite Radar ◽  
Gis Software ◽  
Derived Data

A rapid assessment of the areal extent of landslide disasters is one of the main challenges facing by the scientific community. Satellite radar data represent a powerful tool for the rapid detection of landslides over large spatial scales, even in case of persistent cloud cover. To define landslide locations, radar data need to be firstly pre-processed and then elaborated for the extraction of the required information. Segmentation represents one of the most useful procedures for identifying land cover changes induced by landslides. In this study, we present an application of the i.segment module of GRASS GIS software for segmenting radar-derived data. As study area, we selected the Tagari River valley in Papua New Guinea, where massive landslides were triggered by a M7.5 earthquake on February 25, 2018. A comparison with ground truth data revealed a suitable performance of i.segment. Particular segmentation patterns, in fact, resulted in the areas affected by landslides with respect to the external ones, or to the same areas before the earthquake. These patterns highlighted a relevant contrast of radar backscattering values recorded before and after the landslides. With our procedure, we were able to define the extension of the mass movements that occurred in the study area, just three days after the M7.5 earthquake.

scholarly journals An approach for detecting changes related to natural disasters using Synthetic Aperture Radar data

2015 ◽  
Vol XL-7/W3 ◽  
pp. 819-826 ◽  
Author(s):  
N. Milisavljevic ◽  
D. Closson ◽  
F. Holecz ◽  
F. Collivignarelli ◽  
P. Pasquali
Keyword(s):  
Synthetic Aperture Radar ◽  
Ground Truth ◽  
Radar Data ◽  
Synthetic Aperture ◽  
Optical Remote Sensing ◽  
Ground Truth Data ◽  
Synthetic Aperture Radar Data ◽  
Before And After ◽  
The Impact ◽  
Aperture Radar

Land-cover changes occur naturally in a progressive and gradual way, but they may happen rapidly and abruptly sometimes. Very high resolution remote sensed data acquired at different time intervals can help in analyzing the rate of changes and the causal factors. In this paper, we present an approach for detecting changes related to disasters such as an earthquake and for mapping of the impact zones. The approach is based on the pieces of information coming from SAR (Synthetic Aperture Radar) and on their combination. The case study is the 22 February 2011 Christchurch earthquake. <br><br> The identification of damaged or destroyed buildings using SAR data is a challenging task. The approach proposed here consists in finding amplitude changes as well as coherence changes before and after the earthquake and then combining these changes in order to obtain richer and more robust information on the origin of various types of changes possibly induced by an earthquake. This approach does not need any specific knowledge source about the terrain, but if such sources are present, they can be easily integrated in the method as more specific descriptions of the possible classes. <br><br> A special task in our approach is to develop a scheme that translates the obtained combinations of changes into ground information. Several algorithms are developed and validated using optical remote sensing images of the city two days after the earthquake, as well as our own ground-truth data. The obtained validation results show that the proposed approach is promising.

scholarly journals Validation of Automatic Cb Observations for METAR Messages without Ground Truth

2015 ◽  
Vol 54 (10) ◽  
pp. 2063-2075 ◽  
Author(s):  
Otto Hyvärinen ◽  
Elena Saltikoff ◽  
Harri Hohti
Keyword(s):  
Latent Class Analysis ◽  
Statistical Estimation ◽  
Latent Class ◽  
Detection System ◽  
Ground Truth ◽  
Radar Data ◽  
Convective Rainfall ◽  
Ground Truth Data ◽  
Finnish Meteorological Institute ◽  
Meteorological Institute

AbstractIn aviation meteorology, METAR messages are used to disseminate the existence of cumulonimbus (Cb) clouds. METAR messages are traditionally constructed manually from human observations, but there is a growing trend toward automation of this process. At the Finnish Meteorological Institute (FMI), METAR messages incorporate an operational automatic detection of Cb based solely on weather radar data, when manual observations are not available. However, the verification of this automatic Cb detection is challenging, as good ground truth data are not often available; even human observations are not perfect as Cb clouds can be obscured by other clouds, for example. Therefore, statistical estimation of the relevant verification measures from imperfect observations using latent class analysis (LCA) was explored. In addition to radar-based products and human observations, the convective rainfall rate from EUMETSAT’s Nowcasting Satellite Application Facility and lightning products from the Finnish lightning network were used for determining the existence of Cb clouds. Results suggest that LCA gives reasonable estimates of verification measures and, based on these estimates, the Cb detection system at FMI gives results comparable to human observations.

scholarly journals MONITORING OF WET SNOW AND ACCUMULATIONS AT HIGH ALPINE GLACIERS USING RADAR TECHNOLOGIES

2015 ◽  
Vol XL-7/W3 ◽  
pp. 1063-1068 ◽  
Author(s):  
A. Wendleder ◽  
A. Heilig ◽  
A. Schmitt ◽  
C. Mayer
Keyword(s):  
Mass Balance ◽  
Single Point ◽  
Ground Truth ◽  
Radar Data ◽  
Sar Images ◽  
Ground Truth Data ◽  
Temporal Shift ◽  
Spatial Coverage ◽  
Wet Snow ◽  
Periodic Monitoring

Conventional studies to assess the annual mass balance for glaciers rely on single point observations in combination with model and interpolation approaches. Just recently, airborne and spaceborne data is used to support such mass balance determinations. Here, we present an approach to map temporal changes of the snow cover in glaciated regions of Tyrol, Austria, using SAR-based satellite data. Two dual-polarized SAR images are acquired on 22 and 24 September 2014. As X and C-band reveal different backscattering properties of snow, both TerraSAR-X and RADARSAT-2 images are analysed and compared to ground truth data. Through application of filter functions and processing steps containing a Kennaugh decomposition, ortho-rectification, radiometric enhancement and normalization, we were able to distinguish between dry and wet parts of the snow surface. The analyses reveal that the wet-snow can be unambiguously classified by applying a threshold of -11 dB. Bare ice at the surface or a dry snowpack does not appear in radar data with such low backscatter values. From the temporal shift of wet-snow, a discrimination of accumulation areas on glaciers is possible for specific observation dates. Such data can reveal a periodic monitoring of glaciers with high spatial coverage independent from weather or glacier conditions.

scholarly journals ANALYSIS OF THE GEOMETRY OF SURFACE DEFORMATIONS CAUSED BY INDUCED TREMORS IN THE AREA OF UNDERGROUND COPPER MINING

2020 ◽  
Vol V-3-2020 ◽  
pp. 149-156
Author(s):  
K. Owczarz ◽  
J. Blachowski
Keyword(s):  
Ground Deformation ◽  
Vertical Displacement ◽  
Radar Data ◽  
Seismic Events ◽  
Before And After ◽  
Temporal Intervals ◽  
Vertical Displacements ◽  
Sw Poland ◽  
Satellite Radar ◽  
Continuous Field

Abstract. Induced seismicity by human operations such as mining is usually unpredictable due to the sudden and unexpected character of this phenomenon. The effects of seismic events on the surface, i.e. ground deformation had been difficult to measure with traditional geodetic methods, which are based on discrete point observations and are carried out at temporal intervals and in fixed locations (e.g. levelling lines). Development of radar remote sensing (InSAR) techniques and proliferation of open satellite radar data such as Sentinel- 1 mission provides means that can now be successfully applied to investigate areas and ground movements affected by seismicity induced by mining. In this paper four induced seismic events with magnitudes from 4.5 to 4.8 that occurred between 16 December 2016 and 15 September 2018 in the Rudna underground copper mine area in SW Poland have been investigated with differential satellite radar interferometry (DInSAR). Based on the results of processing of 37 pairs of Sentinel-1 data captured before and after each of these events, deformation areas have been spatially localised and vertical displacement and extent of deformation have been calculated. The mean maximum vertical displacements range from −70 mm for the 4.5 magnitude tremor to −94 mm for the 4.8 magnitude event. Whereas, mean extent ranges from 1.5 km to 1.9 km in the W-E direction and from 1.8 km to 2.1 km in the N-S direction. A linear relation between magnitude of induced tremor and increase in vertical displacement and extent of the ground deformation has been established. Moreover, the results of this study indicate that InSAR is adequately accurate technique to analyse ground displacements caused by mining induced tremors and provides continuous field data on the geometry of the resulting deformation areas.

scholarly journals Assessing Wildfire Burn Severity and Its Relationship with Environmental Factors: A Case Study in Interior Alaska Boreal Forest

2021 ◽  
Vol 13 (10) ◽  
pp. 1966
Author(s):  
Christopher W Smith ◽  
Santosh K Panda ◽  
Uma S Bhatt ◽  
Franz J Meyer ◽  
Anushree Badola ◽  
...  
Keyword(s):  
Boreal Forest ◽  
Ground Truth ◽  
Burn Severity ◽  
Classification Methods ◽  
Spectral Indices ◽  
Ground Truth Data ◽  
Burn Scar ◽  
Interior Alaska ◽  
Remote Sensing Methods ◽  
The Relationship

In recent years, there have been rapid improvements in both remote sensing methods and satellite image availability that have the potential to massively improve burn severity assessments of the Alaskan boreal forest. In this study, we utilized recent pre- and post-fire Sentinel-2 satellite imagery of the 2019 Nugget Creek and Shovel Creek burn scars located in Interior Alaska to both assess burn severity across the burn scars and test the effectiveness of several remote sensing methods for generating accurate map products: Normalized Difference Vegetation Index (NDVI), Normalized Burn Ratio (NBR), and Random Forest (RF) and Support Vector Machine (SVM) supervised classification. We used 52 Composite Burn Index (CBI) plots from the Shovel Creek burn scar and 28 from the Nugget Creek burn scar for training classifiers and product validation. For the Shovel Creek burn scar, the RF and SVM machine learning (ML) classification methods outperformed the traditional spectral indices that use linear regression to separate burn severity classes (RF and SVM accuracy, 83.33%, versus NBR accuracy, 73.08%). However, for the Nugget Creek burn scar, the NDVI product (accuracy: 96%) outperformed the other indices and ML classifiers. In this study, we demonstrated that when sufficient ground truth data is available, the ML classifiers can be very effective for reliable mapping of burn severity in the Alaskan boreal forest. Since the performance of ML classifiers are dependent on the quantity of ground truth data, when sufficient ground truth data is available, the ML classification methods would be better at assessing burn severity, whereas with limited ground truth data the traditional spectral indices would be better suited. We also looked at the relationship between burn severity, fuel type, and topography (aspect and slope) and found that the relationship is site-dependent.

scholarly journals Automatic Evaluation of Wheat Resistance to Fusarium Head Blight Using Dual Mask-RCNN Deep Learning Frameworks in Computer Vision

2020 ◽  
Vol 13 (1) ◽  
pp. 26
Author(s):  
Wen-Hao Su ◽  
Jiajing Zhang ◽  
Ce Yang ◽  
Rae Page ◽  
Tamas Szinyei ◽  
...  
Keyword(s):  
Fusarium Head Blight ◽  
Ground Truth ◽  
Wheat Breeding ◽  
Head Blight ◽  
Detection Rates ◽  
Ground Truth Data ◽  
Resistant Cultivars ◽  
Feature Pyramid ◽  
Rater Error ◽  
Wheat Lines

In many regions of the world, wheat is vulnerable to severe yield and quality losses from the fungus disease of Fusarium head blight (FHB). The development of resistant cultivars is one means of ameliorating the devastating effects of this disease, but the breeding process requires the evaluation of hundreds of lines each year for reaction to the disease. These field evaluations are laborious, expensive, time-consuming, and are prone to rater error. A phenotyping cart that can quickly capture images of the spikes of wheat lines and their level of FHB infection would greatly benefit wheat breeding programs. In this study, mask region convolutional neural network (Mask-RCNN) allowed for reliable identification of the symptom location and the disease severity of wheat spikes. Within a wheat line planted in the field, color images of individual wheat spikes and their corresponding diseased areas were labeled and segmented into sub-images. Images with annotated spikes and sub-images of individual spikes with labeled diseased areas were used as ground truth data to train Mask-RCNN models for automatic image segmentation of wheat spikes and FHB diseased areas, respectively. The feature pyramid network (FPN) based on ResNet-101 network was used as the backbone of Mask-RCNN for constructing the feature pyramid and extracting features. After generating mask images of wheat spikes from full-size images, Mask-RCNN was performed to predict diseased areas on each individual spike. This protocol enabled the rapid recognition of wheat spikes and diseased areas with the detection rates of 77.76% and 98.81%, respectively. The prediction accuracy of 77.19% was achieved by calculating the ratio of the wheat FHB severity value of prediction over ground truth. This study demonstrates the feasibility of rapidly determining levels of FHB in wheat spikes, which will greatly facilitate the breeding of resistant cultivars.

Sign in / Sign up

Export Citation Format

Share Document