A Meta-Methodology for Improving Land Cover and Land Use Classification with SAR Imagery

Per-point classification is a traditional method for remote sensing data classification, and for radar data in particular. Compared with optical data, the discriminative power of radar data is quite limited, for most applications. A way of trying to overcome these difficulties is to use Region-Based Classification (RBC), also referred to as Geographical Object-Based Image Analysis (GEOBIA). RBC methods first aggregate pixels into homogeneous objects, or regions, using a segmentation procedure. Moreover, segmentation is known to be an ill-conditioned problem because it admits multiple solutions, and a small change in the input image, or segmentation parameters, may lead to significant changes in the image partitioning. In this context, this paper proposes and evaluates novel approaches for SAR data classification, which rely on specialized segmentations, and on the combination of partial maps produced by classification ensembles. Such approaches comprise a meta-methodology, in the sense that they are independent from segmentation and classification algorithms, and optimization procedures. Results are shown that improve the classification accuracy from Kappa = 0.4 (baseline method) to a Kappa = 0.77 with the presented method. Another test site presented an improvement from Kappa = 0.36 to a maximum of 0.66 also with radar data.

Download Full-text

POSSIBILITIES OF THE JOINT USE OF OPTICAL AND RADAR DATA IN FLOOD SPACE MONITORING

ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-archives-xlii-3-w4-67-2018 ◽

2018 ◽

Vol XLII-3/W4 ◽

pp. 67-73

Author(s):

O. P. Arkhipkin ◽

G. N. Sagatdinova

Keyword(s):

Spatial Resolution ◽

Complex Analysis ◽

Remote Sensing Data ◽

Weather Conditions ◽

High Water ◽

Radar Data ◽

Time Of Day ◽

Optical Data ◽

Additional Information ◽

Good Spatial Resolution

Abstract. The article gives a brief description of the system of space monitoring of high water and floods. Its main tasks are the operational dynamics of snow and ice cover melting and the passage of flood waters. The solution of these tasks is carried out in three levels corresponding to the low, medium and high resolution of remote sensing data. An important role in monitoring is given to radar data. This is due to the features of the radar survey: independence from weather conditions and time of day, regularity, good spatial resolution, the possibility of using polarimetric properties (including phase information). The use of radar data also provides additional information, including the allocation of wet soils, flooded vegetation and infrastructure. The presence of large time periods of repeated survey, interference (cloudiness, haze, noise, etc.), different spatial resolution necessitates a complex analysis of optical and radar data in flood space monitoring. Such analysis makes it possible to better observe the flood dynamics, more precisely identify of flooding zones and determine their structure. Features of radar survey (transparency of dry snow and change of reflected signal during snowmelt) allow using them to determine the beginning of snow melt and determine the degree of water content in it. Optical data are also used to determine the area and structure of the snow cover. Method of detecting the beginning of the snowmelt period consists in the comparison of the current radar image with a base image created as an average image from the winter images with dry snow.

Download Full-text

Discovering Potential Settlement Areas around Archaeological Tells Using the Integration between Historic Topographic Maps, Optical, and Radar Data in the Northern Nile Delta, Egypt

Remote Sensing ◽

10.3390/rs11243039 ◽

2019 ◽

Vol 11 (24) ◽

pp. 3039 ◽

Cited By ~ 2

Author(s):

Abdelaziz Elfadaly ◽

Mohamed A. R. Abouarab ◽

Radwa R. M. El Shabrawy ◽

Wael Mostafa ◽

Penelope Wilson ◽

...

Keyword(s):

Nile Delta ◽

Remote Sensing Data ◽

Radar Data ◽

Primary Objective ◽

Topographic Maps ◽

Optical Data ◽

History Of ◽

The Creation ◽

Settlement Studies ◽

Settlement Areas

The primary objective of this study is to leverage the integration of surface mapping data derived from optical, radar, and historic topographical studies with archaeological sampling to identify ancient settlement areas in the Northern Nile Delta, Egypt. This study employed the following methods: digitization of topographic maps, band indices techniques on optical data, the creation of a 3D model from SRTM data, and Sentinel-1 interferometric wide swath (IW) analysis. This type of study is particularly relevant to the search for evidence of otherwise hidden ancient settlements. Due to its geographical situation and the fertility of the Nile, Egypt witnessed the autochthonous development of predynastic and dynastic civilizations, as well as an extensive history of external influences due to Greek, Roman, Coptic, Islamic, and Colonial-era interventions. Excavation work at Buto (Tell el-Fara’in) in 2017–18, carried out by the Kafrelsheikh University (KFS) in cooperation with the Ministry of Antiquities, demonstrated that remote sensing data offers considerable promise as a tool for developing regional settlement studies and excavation strategies. This study integrates the mission work in Buto with the satellite imagery in and around the area of the excavation. The results of the initial Buto area research serve as a methodological model to expand the study area to the North Delta with the goal of detecting the extent of the ancient kingdoms of Buto and Sakha. The results of this research include the creation of a composite historical database using ancient references and early topographical maps (1722, 1941, 1950, and 1997), Optical Corona (1965), Landsat MSS (Multispectral Scanner System) (1973, 1978, and 1988), TM (Thematic Mapper) (2005) data, and Radar SRTM (2014) and Sentinel1 (2018 and 2019) data. The data in this study have been analyzed using the ArcMap, Envi, and SNAP software. The results from the current investigation highlight the rapid changes in the land use/land cover in the last century in which many ancient sites were lost due to agriculture and urban development. Three potential settlement areas have been identified with the Sentinel1 Radar data, and have been integrated with the early maps. These discoveries will help develop excavation strategies aimed at elucidating the ancient settlement dynamics and history of the region during the next phase of research.

Download Full-text

Improving the Accuracy of Multiple Algorithms for Crop Classification by Integrating Sentinel-1 Observations with Sentinel-2 Data

Remote Sensing ◽

10.3390/rs13020243 ◽

2021 ◽

Vol 13 (2) ◽

pp. 243

Author(s):

Amal Chakhar ◽

David Hernández-López ◽

Rocío Ballesteros ◽

Miguel A. Moreno

Keyword(s):

Remote Sensing ◽

Vegetation Index ◽

Normalized Difference Vegetation Index ◽

Temporal Dynamics ◽

Remote Sensing Data ◽

Radar Data ◽

Optical Data ◽

Support Vector ◽

Sensing Data ◽

Crop Classification

The availability of an unprecedented amount of open remote sensing data, such as Sentinel-1 and -2 data within the Copernicus program, has boosted the idea of combining the use of optical and radar data to improve the accuracy of agricultural applications such as crop classification. Sentinel-1’s Synthetic Aperture Radar (SAR) provides co- and cross-polarized backscatter, which offers the opportunity to monitor agricultural crops using radar at high spatial and temporal resolution. In this study, we assessed the potential of integrating Sentinel-1 information (VV and VH backscatter and their ratio VH/VV with Sentinel-2A data (NDVI) to perform crop classification and to define which are the most important input data that provide the most accurate classification results. Further, we examined the temporal dynamics of remote sensing data for cereal, horticultural, and industrial crops, perennials, deciduous trees, and legumes. To select the best SAR input feature, we tried two approaches, one based on classification with only SAR features and one based on integrating SAR with optical data. In total, nine scenarios were tested. Furthermore, we evaluated the performance of 22 nonparametric classifiers on which most of these algorithms had not been tested before with SAR data. The results revealed that the best performing scenario was the one integrating VH and VV with normalized difference vegetation index (NDVI) and cubic support vector machine (SVM) (the kernel function of the classifier is cubic) as the classifier with the highest accuracy among all those tested.

Download Full-text

ASSESSMENT OF LANDSLIDE-INDUCED MORPHOLOGY CHANGES USING AN OBJECT-BASED IMAGE ANALYSIS APPROACH: A CASE STUDY OF HÍTARDALUR, ICELAND

ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-archives-xlii-3-w8-109-2019 ◽

2019 ◽

Vol XLII-3/W8 ◽

pp. 109-114

Author(s):

Z. Dabiri ◽

D. Hölbling ◽

L. Abad ◽

D. Tiede

Keyword(s):

Remote Sensing ◽

Image Analysis ◽

Remote Sensing Data ◽

Optical Data ◽

Sensing Data ◽

Object Based Image Analysis ◽

Object Based ◽

Spatio Temporal ◽

The Impact ◽

Morphology Changes

Abstract. On July 7, 2018, a large landslide occurred at the eastern slope of the Fagraskógarfjall Mountain in Hítardalur valley in West Iceland. The landslide dammed the river, led to the formation of a lake and, consequently, to a change in the river course. The main focus of this research is to develop a knowledge-based expert system using an object-based image analysis (OBIA) approach for identifying morphology changes caused by the Hítardalur landslide. We use synthetic aperture radar (SAR) and optical remote sensing data, in particular from Sentinel-1/2 for detection of the landslide and its effects on the river system. We extracted and classified the landslide area, the landslide-dammed lake, other lakes and the river course using intensity information from S1 and spectral information from S2 in the object-based framework. Future research will focus on further developing this approach to support mapping and monitoring of the spatio-temporal dynamics of surface morphology in an object-based framework by combining SAR and optical data. The results can reveal details on the applicability of different remote sensing data for the spatio-temporal investigation of landslides, landslide-induced river course changes and lake formation and lead to a better understanding of the impact of large landslides on river systems.

Download Full-text

BACKSCATTER ANALYSIS USING MULTI-TEMPORAL SENTINEL-1 SAR DATA FOR CROP GROWTH OF MAIZE IN KONYA BASIN, TURKEY

ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-archives-xlii-3-9-2018 ◽

2018 ◽

Vol XLII-3 ◽

pp. 9-13 ◽

Cited By ~ 3

Author(s):

S. Abdikan ◽

A. Sekertekin ◽

M. Ustunern ◽

F. Balik Sanli ◽

R. Nasirzadehdizaji

Keyword(s):

Land Use ◽

Plant Height ◽

Stem Elongation ◽

Test Site ◽

Central Anatolia ◽

Crop Fields ◽

Object Based ◽

Multi Temporal ◽

Sar Imagery ◽

Konya Basin

Temporal monitoring of crop types is essential for the sustainable management of agricultural activities on both national and global levels. As a practical and efficient tool, remote sensing is widely used in such applications. In this study, Sentinel-1 Synthetic Aperture Radar (SAR) imagery was utilized to investigate the performance of the sensor backscatter image on crop monitoring. Multi-temporal C-band VV and VH polarized SAR images were acquired simultaneously by in-situ measurements which was conducted at Konya basin, central Anatolia Turkey. During the measurements, plant height of maize plant was collected and relationship between backscatter values and plant height was analysed. The maize growth development was described under Biologische Bundesanstalt, bundessortenamt und CHemische industrie (BBCH). Under BBCH stages, the test site was classified as leaf development, stem elongation, heading and flowering in general. The correlation coefficient values indicated high correlation for both polarimetry during the early stages of the plant, while late stages indicated lower values in both polarimetry. As a last step, multi-temporal coverage of crop fields was analysed to map seasonal land use. To this aim, object based image classification was applied following image segmentation. About 80&thinsp;% accuracies of land use maps were created in this experiment. As preliminary results, it is concluded that Sentinel-1 data provides beneficial information about plant growth. Dual-polarized Sentinel-1 data has high potential for multi-temporal analyses for agriculture monitoring and reliable mapping.

Download Full-text

Utilization of BEAM and NEST open source toolboxes in education and research

Geoinformatics FCE CTU ◽

10.14311/gi.5.3 ◽

2010 ◽

Vol 5 ◽

pp. 37-48

Author(s):

Markéta Potůčková ◽

Eva Štefanová

Keyword(s):

Remote Sensing ◽

Open Source ◽

Satellite Images ◽

European Space Agency ◽

Radar Data ◽

Optical Data ◽

Basic Principles ◽

Data Formats ◽

Space Agency ◽

Sar Imagery

European Space Agency (ESA) provides several open source toolboxes for visualization, processing and analyzing satellite images acquired both in optical and microwave domains. Basic ERS & Envisat (A)ATSR and MERIS Toolbox (BEAM) was originally developed for easier handling ENVISAT optical data. Today this toolbox supports several raster data formats and datasets collected with other EO instruments such as MODIS, AVHRR, CHRIS/Proba. The NEXT ESA SAR Toolbox (NEST) has been created for processing radar data acquired from different satellites such as ERS 1&2, ENVISAT, RADARSAT or TerraSAR X. Both toolboxes are suitable for the education of the basic principles of data processing (geometric and radiometric corrections, classification, filtering of radar data) but also for research. Possibilities for utilization of these toolboxes in remote sensing courses based on two examples of practical exercises are described. Use of the NEST toolbox is demonstrated on a research project dealing with snow cover detection from SAR imagery.

Download Full-text

CONCEPTUAL FRAMEWORK OF COMBINED PIXEL AND OBJECT-BASED METHOD FOR DELINEATION OF DEBRIS-COVERED GLACIERS

ISPRS Annals of Photogrammetry Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-annals-iv-5-173-2018 ◽

2018 ◽

Vol IV-5 ◽

pp. 173-180 ◽

Cited By ~ 4

Author(s):

R. Sahu ◽

R. D. Gupta

Keyword(s):

High Spatial Resolution ◽

Remote Sensing Data ◽

Optical Data ◽

New Approach ◽

Shadow Area ◽

Object Based ◽

Thermal Data ◽

Automatic Mapping ◽

Glacier Mapping ◽

Manual Correction

Abstract. Delineation of the glacier is an important task for understanding response of glaciers to climate. In Himalayan region, most of the glaciers are covered with debris. Supraglacial debris works as an obstacle for automatic mapping of glacier using remote sensing data. Different methods have been used to reduce this difficulty based on pixel-based and object-based approaches using optical data, thermal data and DEM. Pixel-based glacier mapping is a traditional method for delineation of the glacier but the object-based method has emerged as a new approach in cryosphere application leading to its successful application in different applications. All pixel-based methods require some degree of manual correction because these can’t be delineated automatically, especially in shadow area and debris covered part of the glacier. In the majority of studies, the object-based method has provided higher accuracy to delineate the debris-covered glacier. Spatially high spatial resolution satellite data is best suited for object-based image classification. In future, a combination of pixel-based method and object-based method can be attempted for delineation of the debris-covered glacier along with its critical analysis for suitability. The present paper critically reviews pixel-based and object-based methods as well as provides a framework for combined pixel and object-based method for delineation of debris-covered glacier.

Download Full-text

COMBINATION OF SPECKLE DIVERGENCE AND NEIGHBORHOOD ANALYSIS TO CLASSIFY SETTLEMENT FROM TERASAR-X DATA

International Journal of Remote Sensing and Earth Sciences (IJReSES) ◽

10.30536/j.ijreses.2012.v9.a1820 ◽

2012 ◽

Vol 9 (1) ◽

Author(s):

Rokhis Komarudin ◽

Agung Indrajit

Keyword(s):

Specific Area ◽

Temporal Analysis ◽

Radar Data ◽

Tropical Country ◽

Neighborhood Analysis ◽

Object Based ◽

Radar Imagery ◽

Multi Temporal ◽

Sar Imagery ◽

Spatial Indexes

Abstract. The objectives of this research were to develop and improve methods for determination of settlements area with focus on synthetic aperture radar (SAR) data. Remote sensing settlement classification has made great progress, both for optical and radar data as well for their fusion. Yet, in radar imagery, settlement classification still contains some problems. Several studies on application of radar imagery have been conducted using techniques such as textural analysis, multi-temporal analysis, statistical model, spatial indexes, and object-based classification. Most of the development methods have several problems in the specific area especially in the tropical country. Several studies also showed that settlement classification accuracies were just below 60%. This was not sufficient enough to classify settlement areas using SAR imagery. Therefore, in this research, we proposed a new method i.e., the combination of the speckle divergence and the neighborhood analysis. The proposed method was applied to classify settlement area in Cilacap and Padang Districts of Indonesia. The results showed that the proposed method produced a good accuracy i.e., 85.5% for Cilacap Districts and 78.1% for Padang Districts.

Download Full-text

Comparison of Multi-Resolution Optical Landsat-8, Sentinel-2 and Radar Sentinel-1 Data for Automatic Lineament Extraction: A Case Study of Alichur Area, SE Pamir

Remote Sensing ◽

10.3390/rs11070778 ◽

2019 ◽

Vol 11 (7) ◽

pp. 778 ◽

Cited By ~ 10

Author(s):

Aminov Javhar ◽

Xi Chen ◽

Anming Bao ◽

Aminov Jamshed ◽

Mamadjanov Yunus ◽

...

Keyword(s):

Spatial Resolution ◽

Accuracy Assessment ◽

Remote Sensing Data ◽

Radar Data ◽

Optical Data ◽

Landsat 8 ◽

Geological Investigation ◽

Lineament Extraction ◽

Automatic Lineament Extraction ◽

Sentinel 2A

Lineament mapping, which is an important part of any structural geological investigation, is made more efficient and easier by the availability of optical as well as radar remote sensing data, such as Landsat and Sentinel with medium and high spatial resolutions. However, the results from these multi-resolution data vary due to their difference in spatial resolution and sensitivity to soil occupation. The accuracy and quality of extracted lineaments depend strongly on the spatial resolution of the imagery. Therefore, the aim of this study was to compare the optical Landsat-8, Sentinel-2A, and radar Sentinel-1A satellite data for automatic lineament extraction. The framework of automatic approach includes defining the optimal parameters for automatic lineament extraction with a combination of edge detection and line-linking algorithms and determining suitable bands from optical data suited for lineament mapping in the study area. For the result validation, the extracted lineaments are compared against the manually obtained lineaments through the application of directional filtering and edge enhancement as well as to the lineaments digitized from the existing geological maps of the study area. In addition, a digital elevation model (DEM) has been utilized for an accuracy assessment followed by the field verification. The obtained results show that the best correlation between automatically extracted lineaments, manual interpretation, and the preexisting lineament map is achieved from the radar Sentinel-1A images. The tests indicate that the radar data used in this study, with 5872 and 5865 lineaments extracted from VH and VV polarizations respectively, is more efficient for structural lineament mapping than the Landsat-8 and Sentinel-2A optical imagery, from which 2338 and 4745 lineaments were extracted respectively.

Download Full-text

Mapping boreal peatland ecosystem types from multitemporal radar and optical satellite imagery

Canadian Journal of Forest Research ◽

10.1139/cjfr-2016-0192 ◽

2017 ◽

Vol 47 (4) ◽

pp. 545-559 ◽

Cited By ~ 20

Author(s):

L.L. Bourgeau-Chavez ◽

S. Endres ◽

R. Powell ◽

M.J. Battaglia ◽

B. Benscoter ◽

...

Keyword(s):

Remote Sensing ◽

Remote Sensing Data ◽

Optical Data ◽

Optical Remote Sensing ◽

Fuel Loading ◽

Alos Palsar ◽

Upper Peninsula ◽

Machine Learning Classification ◽

Object Based ◽

Accuracy Assessments

The ability to distinguish peatland types at the landscape scale has implications for inventory, conservation, estimation of carbon storage, fuel loading, and postfire carbon emissions, among others. This paper presents a multisensor, multiseason remote sensing approach to delineate boreal peatland types (wooded bog, open fen, shrubby fen, treed fen) using a combination of multiple dates of L-band (24 cm) synthetic aperture radar (SAR) from ALOS PALSAR, C-band (∼5.6 cm) from ERS-1 or ERS-2, and Landsat 5 TM optical remote sensing data. Imagery was first evaluated over a small test area of boreal Alberta, Canada, to determine the feasibility of using multisensor SAR and optical data to discriminate peatland types. Then object-based and (or) machine-learning classification algorithms were applied to 3.4 million ha of peatland-rich subregions of Alberta, Canada, and the 4.24 million ha region of Michigan’s Upper Peninsula where peatlands are less dominant. Accuracy assessments based on field-sampled sites show high overall map accuracies (93%–94% for Alberta and Michigan), which exceed those of previous mapping efforts.

Download Full-text