Object-Based Time-Constrained Dynamic Time Warping Classification of Crops Using Sentinel-2

Csillik, Ovidiu; Belgiu, Mariana; Asner, Gregory P.; Kelly, Maggi

doi:10.3390/rs11101257

Cited by 81 publications

(64 citation statements)

References 86 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Among the multitude of potential curves, here we show two example cases: (i) with a small negative CC and (ii) with a strong positive CC. For each of the simulated pairs of time series, we estimate five CCs: (i) using the proposed methodology, (ii) using a slotting method with three different bin size (three, seven and 14 days), (iii) DTW with a Sakoe-Chiba band of 45 days (following [8]) and, when possible, using the simultaneous observations only. Figure 3 shows the observed distributions of these estimators.…”

Section: Validation Using Simulationsmentioning

confidence: 99%

“…It consists in warping the time axis for an optimal alignment of the time series. It is possible to constrain the warping to consider only pairs of observations that are not too far from each other, e.g., the Sakoe-Chiba band [8]. The resulting time series share the same length but consist of padded sequences of the same values (when the time series are stretched) so the actual number of genuine observations is smaller.…”

Section: Introductionmentioning

confidence: 99%

“…The resulting time series share the same length but consist of padded sequences of the same values (when the time series are stretched) so the actual number of genuine observations is smaller. Recently, it has been applied to remote sensing applications as well, mainly in image classification (e.g., [8] or [9]). DTW does not directly provide an index of similarity.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Simple Similarity Index for the Comparison of Remotely Sensed Time Series with Scarce Simultaneous Acquisitions

et al. 2019

View full text Add to dashboard Cite

Emergence of new state-of-the-art technologies has enabled an unprecedented amount of high spatial resolution satellite data having great potential for exploitation of extracted time series for a vast range of applications. Despite the high temporal resolution of time series, the number of real observations of optical data that can be utilized is reduced due to meteorological conditions (such as cloud or haze) prevailing at the time of acquisition. This fact has an effect on the density of the retrieved time series and subsequently on a number of coincidental observations when comparing the similarity of time series from two different data sources for which the simultaneous acquisition date is already scarce. Classical tools for assessing the similarity of such time series can prove to be difficult or even impossible because of a lack of simultaneous observations. In this paper, we propose a simple method in order to circumvent this scarcity issue. In the first step, we rely on an interpolation in order to produce artificial time series on the union of the original acquisition dates. Then, we extend the theory of the correlation coefficient (CC) estimator to these interpolated time series. After validation on synthetic data, this simple approach proved to be extremely efficient on a real case study where Sentinel-2 and PlanetScope NDVI time series on parcels in The Netherlands are compared. Indeed, compared to other methods, it reduced the number of undecided cases while also improving the power of the statistical test on the similarity between both types of time series and the precision of the estimated CC.

show abstract

Section: Validation Using Simulationsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Simple Similarity Index for the Comparison of Remotely Sensed Time Series with Scarce Simultaneous Acquisitions

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Dynamic Time Warping (DTW) is an efficient solution for a non-linear alignment of two temporal sequences (Sakoe, Chiba, 1978). DTW obtained good crops classification results with a reduced number of training samples and with multi-temporal images that present gaps caused by the presence of clouds (Belgiu, Csillik, 2018;Csillik et al, 2019;Guan et al, 2018;Maus et al, 2016;Petitjean, et al, 2011). DTW has initially been developed for speech recognition applications (Sakoe, Chiba, 1978), and it has been adopted in other domains including medicine (Tsevas, Iakovidis, 2010) or robotics (Johnen, Kuhlenkoetter, 2016).…”

Section: Introductionmentioning

confidence: 99%

Dynamic Time Warping for Crops Mapping

Belgiu

Zhou

Marshall

et al. 2020

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Abstract. Dynamic Time Warping (DTW) has been successfully used for crops mapping due to its capability to achieve good classification results when a reduced number of training samples and irregular satellite image time series is available. Despite its recognized advantages, DTW does not account for the duration and seasonality of crops and local differences when assessing the similarity between two temporal sequences. In this study, we implemented a Weighted Derivative modification of DTW (WDDTW) and compared it with DTW and Time Weighted Dynamic Time Warping (TWDTW) for crops mapping. We show that WDDTW outperformed DTW achieving an overall accuracy of 67 %, whereas DTW obtained an accuracy of 57%. Yet, TWDTW performed better than both methods obtaining an accuracy of 88%.

show abstract

“…While in the pixel-based approach the reference for classification is the individual value of each pixel, without taking into account the neighborhood influence in the decision process, GEOBIA is characterized by the increase of information related to objects, such as shape, texture, compacity and context, which is useful to analyze heterogeneous landscapes (BLASCHKE, 2010). Recent studies have reported the advantages of detecting landscape changes via GEOBIA because it considers key factors that transcend reflectance values (SCHULTZ et al, 2016;CSILLIK, 2018;CSILLIK et al, 2019), outperforming pixel-based classification methods. In this context, considering that the generation of geo-objects may favor the detection of clearings in new frontiers of deforestation, such as Rondon do Pará, we adopted the objectoriented GEOBIA approach to perform land use and land cover (LULC) classifications in the present study.…”

Section: Introductionmentioning

confidence: 99%

Land Use and Land Cover Mapping in a Priority Municipality for Deforestation Control Actions in the Amazon using GEOBIA

Conceição¹,

Chaves²,

Mataveli³

2020

Rev. Bras. Cartogr.

View full text Add to dashboard Cite

The objective of this work was to detect land use and land cover changes in the municipality of Rondon do Pará, in the Brazilian Legal Amazon, through the Geographic Object-based Image Analysis (GEOBIA), aiming to identify whether its inclusion in the list of priority municipalities for actions to prevent, monitor and control deforestation in the Amazon (Decree 6.321/2007), has inhibited this process. The methodological procedure involved the calculation of vegetation indices to highlight interest classes, the segmentation of the images into geo-objects and the classification using Support Vector Machine (SVM) algorithm. Also, we evaluated the relationship between annual rates of particulate material with a diameter of fewer than 2.5 micrometers (PM2.5) emitted from fires and deforestation in Rondon do Pará. The results show the reduction of deforestation rates but the conversion, in absolute values, of 984,47 km² of natural vegetation to anthropized areas along the 2006-2017 period, and the reduction of the average PM2.5 fire emissions from 31,107 tons year-1 between 2002 and 2007 to 10,394 ton year-1 in the post-inclusion period (2008-2019). This represents that the inclusion of Rondon do Pará in this list was beneficial, but the suppression of natural vegetation was not totally curbed, reinforcing that changes from environmental policies are gradual. Activities that do not use fire to suppress natural vegetation, such as selective logging, common in Rondon do Pará, had impact. This makes it necessary to intensify inspections and the development of sustainable actions in the region.

show abstract

Object-Based Time-Constrained Dynamic Time Warping Classification of Crops Using Sentinel-2

Cited by 81 publications

References 86 publications

A Simple Similarity Index for the Comparison of Remotely Sensed Time Series with Scarce Simultaneous Acquisitions

A Simple Similarity Index for the Comparison of Remotely Sensed Time Series with Scarce Simultaneous Acquisitions

Dynamic Time Warping for Crops Mapping

Land Use and Land Cover Mapping in a Priority Municipality for Deforestation Control Actions in the Amazon using GEOBIA

Contact Info

Product

Resources

About