Abstract
This paper studies the detection of anomalous crop development at the parcel-level based on an unsupervised outlier detection technique. The experimental validation is conducted on rapeseed and wheat parcels located in Beauce (France). The proposed methodology consists of four sequential steps: (1) preprocessing of synthetic aperture radar (SAR) and multispectral images acquired using Sentinel-1 and Sentinel-2 satellites, (2) extraction of SAR and multispectral pixel-level features, (3) computation of parcel-level features using zonal statistics and (4) outlier detection. The different types of anomalies that can affect the studied crops are analyzed and described. The different factors that can influence the outlier detection results are investigated with a particular attention devoted to the synergy between Sentinel-1 and Sentinel-2 data. Overall, the best performance is obtained when using jointly a selection of Sentinel-1 and Sentinel-2 features with the isolation forest algorithm. The selected features are co-polarized (VV) and cross-polarized (VH) backscattering coefficients for Sentinel-1 and five Vegetation Indexes for Sentinel-2 (among us, the Normalized Difference Vegetation Index and two variants of the Normalized Difference Water). When using these features with an outlier ratio of 10%, the percentage of detected true positives (i.e., crop anomalies) is equal to 94.1% for rapeseed parcels and 95.5% for wheat parcels.
Highlights
Monitoring crop growth and status is a major challenge in remote sensing for agriculture [1]
This paper studied a new anomaly detection method for crop monitoring based on outlier analysis at the parcel-level using Sentinel-1 and Sentinel-2 features
This method is decomposed into 4 main steps: (1) preprocessing of multispectral and synthetic aperture radar (SAR) images, (2) computation of pixel-level features, (3) computation of zonal statistics at the parcel-level for all pixel-level features at each date, (4) detection of abnormal crop parcels using the isolation forest algorithm with the multi-temporal zonal statistics
Summary
Monitoring crop growth and status is a major challenge in remote sensing for agriculture [1]. Images have been widely studied since they are available regardless of sunlight and cloud coverage conditions [6,7,8,9,10]. The complementarity of these two types of images has been used to address problems including crop type classification [11,12], estimation of crop water requirement [13] and change detection [14,15,16]. The joint use of SAR and multispectral images is encouraged by the large amount of free data provided by the Sentinel-1. Various studies have introduced methodologies for deriving crop classification maps using S1 and
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
