scholarly journals Ice Monitoring in Swiss Lakes from Optical Satellites and Webcams Using Machine Learning

2020 ◽  
Vol 12 (21) ◽  
pp. 3555
Author(s):  
Manu Tom ◽  
Rajanie Prabha ◽  
Tianyu Wu ◽  
Emmanuel Baltsavias ◽  
Laura Leal-Taixé ◽  
...  
Keyword(s):  
Machine Learning ◽  
Satellite Images ◽  
Global Climate ◽  
Semantic Segmentation ◽  
Lake Ice ◽  
Climate Indicators ◽  
Multi Temporal ◽  
Spatio Temporal ◽  
Set Up ◽  
Optical Satellite Images

Continuous observation of climate indicators, such as trends in lake freezing, is important to understand the dynamics of the local and global climate system. Consequently, lake ice has been included among the Essential Climate Variables (ECVs) of the Global Climate Observing System (GCOS), and there is a need to set up operational monitoring capabilities. Multi-temporal satellite images and publicly available webcam streams are among the viable data sources capable of monitoring lake ice. In this work we investigate machine learning-based image analysis as a tool to determine the spatio-temporal extent of ice on Swiss Alpine lakes as well as the ice-on and ice-off dates, from both multispectral optical satellite images (VIIRS and MODIS) and RGB webcam images. We model lake ice monitoring as a pixel-wise semantic segmentation problem, i.e., each pixel on the lake surface is classified to obtain a spatially explicit map of ice cover. We show experimentally that the proposed system produces consistently good results when tested on data from multiple winters and lakes. Our satellite-based method obtains mean Intersection-over-Union (mIoU) scores > 93%, for both sensors. It also generalises well across lakes and winters with mIoU scores > 78% and >80% respectively. On average, our webcam approach achieves mIoU values of ≈87% and generalisation scores of ≈71% and ≈69% across different cameras and winters respectively. Additionally, we generate and make available a new benchmark dataset of webcam images (Photi-LakeIce) which includes data from two winters and three cameras.

scholarly journals LAKE ICE DETECTION IN LOW-RESOLUTION OPTICAL SATELLITE IMAGES

2018 ◽  
Vol IV-2 ◽  
pp. 279-286 ◽  
Author(s):  
M. Tom ◽  
U. Kälin ◽  
M. Sütterlin ◽  
E. Baltsavias ◽  
K. Schindler
Keyword(s):  
Satellite Images ◽  
Visual Analysis ◽  
Semantic Segmentation ◽  
Support Vector ◽  
Lake Ice ◽  
Processing Step ◽  
Multi Temporal ◽  
Natural Data ◽  
Ice Detection ◽  
Optical Satellite Images

Monitoring and analyzing the (decreasing) trends in lake freezing provides important information for climate research. Multi-temporal satellite images are a natural data source to survey ice on lakes. In this paper, we describe a method for lake ice monitoring, which uses low spatial resolution (250 m–1000 m) satellite images to determine whether a lake is frozen or not. We report results on four selected lakes in Switzerland: Sihl, Sils, Silvaplana and St. Moritz. These lakes have different properties regarding area, altitude, surrounding topography and freezing frequency, describing cases of medium to high difficulty. Digitized Open Street Map (OSM) lake outlines are back-projected on to the image space after generalization. As a pre-processing step, the absolute geolocation error of the lake outlines is corrected by matching the projected outlines to the images. We define the lake ice detection as a two-class (frozen, non-frozen) semantic segmentation problem. Several spectral channels of the multi-spectral satellite data are used, both reflective and emissive (thermal). Only the cloud-free (clean) pixels which lie completely inside the lake are analyzed. The most useful channels to solve the problem are selected with xgboost and visual analysis of histograms of reference data, while the classification is done with non-linear support vector machine (SVM). We show experimentally that this straight-forward approach works well with both MODIS and VIIRS satellite imagery. Moreover, we show that the algorithm produces consistent results when tested on data from multiple winters.

scholarly journals LAKE ICE DETECTION FROM SENTINEL-1 SAR WITH DEEP LEARNING

2020 ◽  
Vol V-3-2020 ◽  
pp. 409-416 ◽  
Author(s):  
M. Tom ◽  
R. Aguilar ◽  
P. Imhof ◽  
S. Leinss ◽  
E. Baltsavias ◽  
...  
Keyword(s):  
Global Climate ◽  
Semantic Segmentation ◽  
Limiting Factor ◽  
Lake Ice ◽  
Convolutional Network ◽  
Important Indicator ◽  
Proposed Model ◽  
Spatio Temporal ◽  
Ice Detection ◽  
Validation Tests

Abstract. Lake ice, as part of the Essential Climate Variable (ECV) lakes, is an important indicator to monitor climate change and global warming. The spatio-temporal extent of lake ice cover, along with the timings of key phenological events such as freeze-up and break-up, provide important cues about the local and global climate. We present a lake ice monitoring system based on the automatic analysis of Sentinel-1 Synthetic Aperture Radar (SAR) data with a deep neural network. In previous studies that used optical satellite imagery for lake ice monitoring, frequent cloud cover was a main limiting factor, which we overcome thanks to the ability of microwave sensors to penetrate clouds and observe the lakes regardless of the weather and illumination conditions. We cast ice detection as a two class (frozen, non-frozen) semantic segmentation problem and solve it using a state-of-the-art deep convolutional network (CNN).We report results on two winters (2016–17 and 2017–18) and three alpine lakes in Switzerland. The proposed model reaches mean Intersection-over-Union (mIoU) scores >90% on average, and >84% even for the most difficult lake. Additionally, we perform cross-validation tests and show that our algorithm generalises well across unseen lakes and winters.

scholarly journals Geoinformatics Based Sprawl Dynamics Analysis of Berhampur Development Authority Area, Odisha, India

2019 ◽  
Vol 8 (4) ◽  
pp. 10471-10477
Keyword(s):  
Urban Areas ◽  
Satellite Images ◽  
Spatial Information ◽  
Spatial Dimension ◽  
High Resolution Satellite Images ◽  
Multi Temporal ◽  
Authority Area ◽  
Arc Gis ◽  
Spatio Temporal ◽  
Gis Software

Urban and Regional planners need accurate and authentic spatio-temporal information of urban sprawls for efficient and sustainable planning of towns & cities worldwide. Geoinformatics powered with temporal high resolution satellite images, Geographic Information System (GIS), mobile technology, etc is now emerged as the most powerful tool for mapping and monitoring the sprawls of urban habitations. In this paper an attempt is made for analysing the dynamics of sprawls of three statutory towns of Berhampur Development Authority (BeDA) area of Ganjam District, Odisha state, India. The spatial information of urban sprawl of each town has been generated using openly available toposheets and multi -sensor & multi - temporal satellite images and the spatio temporal characteristics of sprawls has been analysed in Arc GIS software. The sprawl area as well as the population of the three towns have been analysed and the future scenario of sprawl-population dynamics has been forecasted for the years 2021 and 2031.The result of this paper highlights that sprawls of the three towns i.e Berhampur, Chhatrapur and Gopalpur will expand their spatial dimension by 22,18 and 97 percent by 2031 whereas population of the three towns will increase by 43, 19 and 15 percent between 2011 -2031.Finally the result indicates that there will be decrease in population density in the three towns which will ultimately force the Development Authority to plan more basic infrastructures and transportation in the newly expanded urban areas.

scholarly journals Ship Detection in Optical Satellite Images via Directional Bounding Boxes Based on Ship Center and Orientation Prediction

2019 ◽  
Vol 11 (18) ◽  
pp. 2173 ◽  
Author(s):  
Jinlei Ma ◽  
Zhiqiang Zhou ◽  
Bo Wang ◽  
Hua Zong ◽  
Fei Wu
Keyword(s):  
Remote Sensing ◽  
Satellite Images ◽  
Detection Method ◽  
Semantic Segmentation ◽  
Detection Methods ◽  
Optical Remote Sensing ◽  
Ship Detection ◽  
Bounding Boxes ◽  
Typical Object ◽  
Optical Satellite Images

To accurately detect ships of arbitrary orientation in optical remote sensing images, we propose a two-stage CNN-based ship-detection method based on the ship center and orientation prediction. Center region prediction network and ship orientation classification network are constructed to generate rotated region proposals, and then we can predict rotated bounding boxes from rotated region proposals to locate arbitrary-oriented ships more accurately. The two networks share the same deconvolutional layers to perform semantic segmentation for the prediction of center regions and orientations of ships, respectively. They can provide the potential center points of the ships helping to determine the more confident locations of the region proposals, as well as the ship orientation information, which is beneficial to the more reliable predetermination of rotated region proposals. Classification and regression are then performed for the final ship localization. Compared with other typical object detection methods for natural images and ship-detection methods, our method can more accurately detect multiple ships in the high-resolution remote sensing image, irrespective of the ship orientations and a situation in which the ships are docked very closely. Experiments have demonstrated the promising improvement of ship-detection performance.

Modeling Spatio-Temporal Shoreline Shifting of a Coastal Island in Bangladesh Using Geospatial Techniques and DSAS Extension

2018 ◽  
Vol 18 (1) ◽  
pp. 1-13 ◽  
Author(s):  
Md. Salauddin ◽  
Khandaker Tanvir Hossain ◽  
Istiaqe Ahmed Tanim ◽  
Md. Anisul Kabir ◽  
Mehedi Hasan Saddam
Keyword(s):  
Satellite Images ◽  
River Estuary ◽  
Northwestern Part ◽  
Southwestern Part ◽  
Geospatial Techniques ◽  
End Point ◽  
Time Period ◽  
Multi Temporal ◽  
Spatio Temporal ◽  
End Point Rate

Abstract The study attempted to assess the changes in shoreline and erosion-accretion of newly formed island at the mouth of the Meghna river estuary in Bangladesh using multi temporal satellite images and GIS techniques. The current study used NDWI an MNDWI to delineate land and water boundary to extract the shoreline and also used some overlay analysis to measure the erosion-accretion. DSAS extension is used for analyzing the shifting of the shoreline. The results (1990-2015) show that the island has 1192 hectares of land accreted during this time period, and about 1 km of its shoreline lost during this time period as it has broken in few places. Most accretion found in the northwestern part and erosion in southeastern and southwestern part. The study area was divided into four different segments and about 115 transects were constructed, of which about 74 shows the seaward movement and 44 shows landward movement. Seaward movement and rate of shoreline shifting is higher in the northwestern part where net shoreline movement (NSM) is +1897 meters and end point rate (EPR) is +73 m/year. Landward movement and the rate of shoreline shifting are higher in southeastern and southwestern part of which net shoreline movement (NSM) is about -2680 meters and end point rate (EPR) is about -129 m/year. The highest landward movement is found as -2680 meters and highest seaward movement as +1897 meters. Accretion process is dominant while erosion process discontinued the shoreline in some places.

scholarly journals Monitoring and Landscape Quantification of Uncontrolled Urbanisation in Oasis Regions: The Case of Adrar City in Algeria

2021 ◽  
Vol 5 (2) ◽  
pp. 209-219
Author(s):  
Assoule Dechaicha ◽  
Adel Daikh ◽  
Djamel Alkama
Keyword(s):  
Landscape Metrics ◽  
Satellite Images ◽  
Methodological Approach ◽  
Monitoring And Evaluation ◽  
Temporal Analysis ◽  
Multi Temporal ◽  
Spatio Temporal ◽  
Landsat Satellite ◽  
Landscape Transformations ◽  
The City

Nowadays, uncontrolled urbanisation is one of the major problems facing Algerian oasis regions. The monitoring and evaluation of its landscape transformations remain a key step for any oasis sustainability project. This study highlights the evolution of spatial growth in the city of Adrar in southern Algeria during the period 1986-2016 by establishing a Spatio-temporal mapping and landscape quantification. The methodological approach is based on a multi-temporal analysis of Landsat satellite images for 1986, 1996, 2006 and 2016, and the application of landscape metrics. The results show two opposite spatial trends: significant growth of built-up areas against an excessive loss of palm groves. The landscape metrics allowed the identification of a progressive fragmentation process characterising the palm groves. Thus, the findings of this study show the utility of satellite imagery and landscape metrics approach for monitoring urbanisation patterns and assessing their impacts on oasis ecosystems.

Sign in / Sign up

Export Citation Format

Share Document