Classification of rainfall radar images using the scattering transform

Lagrange, Mathieu; Andrieu, Hervé; Emmanuel, Isabelle; Busquets, Gerard; Loubrié, Stéphane

doi:10.1016/j.jhydrol.2016.06.063

Cited by 18 publications

(15 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The LBP(x, y) of each pixel inside this patch are concatenated to create a fingerprint of the local texture around the pixel at the center of the patch. Equations (5) and (6) are applied on all patches of an image.…”

Section: Other Methodsmentioning

confidence: 99%

“…Despite its intrinsic interest to address multiple scales problems compared to deep learning, scatter transform since its introduction in 2013 has been applied only on a relatively small variety of pattern recognition computer vision problems notably including iris recognition [4], rainfall classification in radar images [5], cell-scale characterization [6,7], or face recognition, [8]. Also, in these applications scatter transform has shown its efficiency, but it was not systematically compared with other techniques in a comprehensible way.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Supervised Image Classification by Scattering Transform with Application to Weed Detection in Culture Crops of High Density

et al. 2019

View full text Add to dashboard Cite

In this article, we assess the interest of the recently introduced multiscale scattering transform for texture classification applied for the first time in plant science. Scattering transform is shown to outperform monoscale approaches (gray-level co-occurrence matrix, local binary patterns) but also multiscale approaches (wavelet decomposition) which do not include combinatory steps. The regime in which scatter transform also outperforms a standard CNN architecture in terms of data-set size is evaluated (10 4 instances). An approach on how to optimally design the scatter transform based on energy contrast is provided. This is illustrated on the hard and open problem of weed detection in culture crops of high density from the top view in intensity images. An annotated synthetic data-set available under the form of a data challenge and a simulator are proposed for reproducible science. Scatter transform only trained on synthetic data shows an accuracy of 85% when tested on real data.3 of 16 the weeds included in the data-set Figure 3 are much more indented. Therefore, this computer vision problem is well adapted to test scatter transform which is a texture-based technique. Camera High density cropRemote Sens. 2019, 11, 249 7 of 16 no other anatomical items larger than the leaf itself. The number of layers tested was up to M = 4 as proposed in [3] since the energy after some layers although none zero is logically vanishing.Interestingly, these specific properties of the scatter transform match the intrinsic multiscale textural 127 nature of our weed detection problem which therefore constitutes an appropriate use case to assess 128 the potential of the scatter transform in practice. A visualization of output images for various filter 129 scale j at m = 2 for a given orientation is shown in Fig. 7. It clearly appears in Fig. 7 that the various 130 scales (texture of the limb and veins at j=3, border shape at j=4 and global leaf shape at j=8-not shown) 131 presented in section 2.2 can be captured with the different scaling factor applied on the wavelet. In our 132 study, we empirically picked L = 8 orientations and investigated up to J = 8 scales since there are 133 no other anatomical items larger than the leaf itself. The number of layers tested was up to M = 4 as 134 proposed in [3] since the energy after some layers although none zero is logically vanishing. 135

show abstract

Section: Other Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Supervised Image Classification by Scattering Transform with Application to Weed Detection in Culture Crops of High Density

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Recently, considerable attention has been paid to increasing the resolution of stochastic rainfall models so that they can mimic rainfall at sub-daily timescales. Currently, several high-resolution stochastic rainfall models are able to deal with precipitation data at typical resolutions of 1 min to 1 h in time and of 100×100 m 2 to 1×1 km 2 in space (see for example Leblois and Creutin, 2013;Paschalis et al, 2013;Benoit et al, 2018). At such scales, not only the marginal distribution of observed rain intensity matters but the space-time dependencies within rain fields are also important features of the rain process (Emmanuel et al, 2012;Marra and Morin, 2018).…”

Section: Introductionmentioning

confidence: 99%

Dealing with non-stationarity in sub-daily stochastic rainfall models

Benoît

Vrac

Mariethoz

2018

Hydrol. Earth Syst. Sci.

View full text Add to dashboard Cite

Abstract. Understanding the stationarity properties of rainfall is critical when using stochastic weather generators. Rainfall stationarity means that the statistics being accounted for remain constant over a given period, which is required for both inferring model parameters and simulating synthetic rainfall. Despite its critical importance, the stationarity of precipitation statistics is often regarded as a subjective choice whose examination is left to the judgement of the modeller. It is therefore desirable to establish quantitative and objective criteria for defining stationary rain periods. To this end, we propose a methodology that automatically identifies rain types with homogeneous statistics. It is based on an unsupervised classification of the space–time–intensity structure of weather radar images. The transitions between rain types are interpreted as non-stationarities. Our method is particularly suited to deal with non-stationarity in the context of sub-daily stochastic rainfall models. Results of a synthetic case study show that the proposed approach is able to reliably identify synthetically generated rain types. The application of rain typing to real data indicates that non-stationarity can be significant within meteorological seasons, and even within a single storm. This highlights the need for a careful examination of the temporal stationarity of precipitation statistics when modelling rainfall at high resolution.

show abstract

“…To model rain non-stationarity on a more data-driven basis, and thereby account for the sub-daily non-stationarities reported above, it has recently been proposed to classify rain fields into rain types (e.g. based on weather radar images) prior to stochastic modelling (Lagrange et al, 2018;Benoit et al, 2018b). Rain fields belonging to the same rain type are then deemed statistically similar, and periods with a constant rain type can be regarded as stationary periods for the simulation of 20 rain intensity.…”

mentioning

confidence: 99%

Accounting for rain type non-stationarity in sub-daily stochastic weather generators

Benoît

Vrac

Mariethoz

2019

Preprint

View full text Add to dashboard Cite

At sub-daily resolution, rain intensity exhibits a strong variability in space and time, which is favorably modelled using stochastic approaches. This strong variability is further enhanced because of the diversity of processes that produce rain (e.g. frontal storms, mesoscale convective systems, local convection), which results in a multiplicity of space-time patterns embedded into rain fields, and in turn leads to non-stationarity of rain statistics. To account for this non-stationarity in the context of stochastic weather generators, and therefore preserve the climatological coherence of rain simulations, we propose 5 to resort to rain types simulation.We explore two methods to simulate rain type time series conditional to meteorological covariates: a parametric approach based on a non-homogeneous semi-Markov chain, and a non-parametric approach based on multiple-point statistics. Both methods are tested by cross-validation using a 17-year long rain type time series defined over central Germany. Evaluation results indicate that the non-parametric approach better simulates the relationships between rain types and meteorological covariates. 10Indeed, the inherent simplifications in the parametric model do not allow fully resolving complex and non-linear interactions between the rainfall statistics and meteorological covariates.The proposed approach is applied to generate rain type time series conditional to meteorological covariates simulated by a Regional Climate Model under an RCP8.5 emission scenario. Results indicate that, by the end of the century, the distribution of rain types could be modified over the area of interest, with an increased frequency of convective-and frontal-like rains at 15 the expense of more stratiform events.

show abstract

Classification of rainfall radar images using the scattering transform

Cited by 18 publications

References 27 publications

Supervised Image Classification by Scattering Transform with Application to Weed Detection in Culture Crops of High Density

Supervised Image Classification by Scattering Transform with Application to Weed Detection in Culture Crops of High Density

Dealing with non-stationarity in sub-daily stochastic rainfall models

Accounting for rain type non-stationarity in sub-daily stochastic weather generators

Contact Info

Product

Resources

About