The Earth Observation-based Anomaly Detection (EOAD) system: A simple, scalable approach to mapping in-field and farm-scale anomalies using widely available satellite imagery

Villamor, Liliana Carolina Castillo; Hardy, Andrew; Bunting, Peter; Llanos-Peralta, Willian; Zamora, Miguel A.; Rodríguez, Y. Chao; Latorre, Douglas Andrés Gómez

doi:10.1016/j.jag.2021.102535

Cited by 4 publications

(4 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Afterward, several thresholds were calculated based on the true and false positive rates. Anomaly detection on crop sites was also performed by [57], where the authors used a histogram to identify the anomalous pixels. In [58], the authors also integrated an outlier detection step during their dataset preparation stage to target the use case of grassland mowing detection over Estonia.…”

Section: Anomaly Detectionmentioning

confidence: 99%

AgriSen-COG, a Multicountry, Multitemporal Large-Scale Sentinel-2 Benchmark Dataset for Crop Mapping Using Deep Learning

Selea

2023

Remote Sensing

View full text Add to dashboard Cite

With the increasing volume of collected Earth observation (EO) data, artificial intelligence (AI) methods have become state-of-the-art in processing and analyzing them. However, there is still a lack of high-quality, large-scale EO datasets for training robust networks. This paper presents AgriSen-COG, a large-scale benchmark dataset for crop type mapping based on Sentinel-2 data. AgriSen-COG deals with the challenges of remote sensing (RS) datasets. First, it includes data from five different European countries (Austria, Belgium, Spain, Denmark, and the Netherlands), targeting the problem of domain adaptation. Second, it is multitemporal and multiyear (2019–2020), therefore enabling analysis based on the growth of crops in time and yearly variability. Third, AgriSen-COG includes an anomaly detection preprocessing step, which reduces the amount of mislabeled information. AgriSen-COG comprises 6,972,485 parcels, making it the most extensive available dataset for crop type mapping. It includes two types of data: pixel-level data and parcel aggregated information. By carrying this out, we target two computer vision (CV) problems: semantic segmentation and classification. To establish the validity of the proposed dataset, we conducted several experiments using state-of-the-art deep-learning models for temporal semantic segmentation with pixel-level data (U-Net and ConvStar networks) and time-series classification with parcel aggregated information (LSTM, Transformer, TempCNN networks). The most popular models (U-Net and LSTM) achieve the best performance in the Belgium region, with a weighted F1 score of 0.956 (U-Net) and 0.918 (LSTM).The proposed data are distributed as a cloud-optimized GeoTIFF (COG), together with a SpatioTemporal Asset Catalog (STAC), which makes AgriSen-COG a findable, accessible, interoperable, and reusable (FAIR) dataset.

show abstract

Section: Anomaly Detectionmentioning

confidence: 99%

AgriSen-COG, a Multicountry, Multitemporal Large-Scale Sentinel-2 Benchmark Dataset for Crop Mapping Using Deep Learning

Selea

2023

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…The geographic coordinates of points (in other words, spatial data components) can serve as another effective predictor of forage biomass productivity [9,5].…”

Section: Construction Of a 3d Model Of The Considered Pasture Areamentioning

confidence: 99%

“…This model enables the estimation of surface topography, moisture accumulation, as well as the accumulation forage and pasture plant biomass. The Earth Observation-based Anomaly Detection (EOAD) approach presented in an article by Liliana Castillo-Villamor et al was able to map plot-level anomalies in rice crops in Colombia, while providing a plan for their elimination [5].…”

Section: Fig 1 Color Interpretation Of Elevations (Cell Width In Meters)mentioning

confidence: 99%

Application of remote monitoring and ground-based sensing in pastoralism

Oleinik,

Skripkin,

Lesnyak

et al. 2023

BIO Web Conf.

View full text Add to dashboard Cite

The development and implementation of an innovative system applying digital aerospace technologies in pastoralism constitute one of the current trends in agricultural development designed to solve numerous problems associated with soil fertility restoration in grazing pastures. Here, a promising line of research is to explore the feasibility of using satellite systems for an overall nutrient yield assessment per hectare of grazing land. The present article describes a comprehensive approach to the optimization of pastoralism that is based on remote methods for assessing pasture fertility using unmanned aerial vehicles (UAVs) and artificial Earth satellites. An analysis of existing methodological approaches reveals that the formalization of prediction processes is complicated by the lack of a theoretical basis for creating appropriate model-algorithmic support. The data on vegetation index dynamics and the nutritional values of forage plants obtained by interpreting imagery from a UAV camera and the multispectral cameras of a satellite service, as well as data from a portable handheld nitrogen sensor, were compared with the actual nutritional values of pasture plants. The study results provide a means to optimize the grouping of grazing animals, taking into account the actual possibility of achieving an increase in live weight. The provided findings indicate the possibility of achieving an additional 11.06% increase in daily live weight gain in young sheep (Jalgin Merino) when keeping them in pasture areas having a vegetation index of at least 0.5. Remote monitoring based on satellite service allows more efficient use of pastures. Study shows a positive relationship between remote sensing NDVI and feed nutritional value. Animal grazing optimization provides an additional 11.06% increase in live weight gain.

show abstract

“…Anomaly detection has been applied in several domains, such as fraud detection, medical imaging, Internet of Things (IoT), surveillance and monitoring and time series data analysis [ 8 ]. In the agricultural domain, it has been mostly applied in precision farming [ 9 , 10 , 11 , 12 , 13 ] and to a far lesser extent in navigation and obstacle detection [ 14 , 15 ].…”

Section: Related Workmentioning

confidence: 99%

Anomaly Detection for Agricultural Vehicles Using Autoencoders

Mujkic

Philipsen

Moeslund

et al. 2022

Sensors

View full text Add to dashboard Cite

The safe in-field operation of autonomous agricultural vehicles requires detecting all objects that pose a risk of collision. Current vision-based algorithms for object detection and classification are unable to detect unknown classes of objects. In this paper, the problem is posed as anomaly detection instead, where convolutional autoencoders are applied to identify any objects deviating from the normal pattern. Training an autoencoder network to reconstruct normal patterns in agricultural fields makes it possible to detect unknown objects by high reconstruction error. Basic autoencoder (AE), vector-quantized variational autoencoder (VQ-VAE), denoising autoencoder (DAE) and semisupervised autoencoder (SSAE) with a max-margin-inspired loss function are investigated and compared with a baseline object detector based on YOLOv5. Results indicate that SSAE with an area under the curve for precision/recall (PR AUC) of 0.9353 outperforms other autoencoder models and is comparable to an object detector with a PR AUC of 0.9794. Qualitative results show that SSAE is capable of detecting unknown objects, whereas the object detector is unable to do so and fails to identify known classes of objects in specific cases.

show abstract

The Earth Observation-based Anomaly Detection (EOAD) system: A simple, scalable approach to mapping in-field and farm-scale anomalies using widely available satellite imagery

Cited by 4 publications

References 45 publications

AgriSen-COG, a Multicountry, Multitemporal Large-Scale Sentinel-2 Benchmark Dataset for Crop Mapping Using Deep Learning

AgriSen-COG, a Multicountry, Multitemporal Large-Scale Sentinel-2 Benchmark Dataset for Crop Mapping Using Deep Learning

Application of remote monitoring and ground-based sensing in pastoralism

Anomaly Detection for Agricultural Vehicles Using Autoencoders

Contact Info

Product

Resources

About