Anomalies detection and cause analysis of autumn crops in individual croplands using timeseries of Sentinel-2 satellite data (Case Study: Golestan province)

Document Type : Research Paper


1 Department of Photogrammetry & Remote Sensing, Faculty of Geomatics Engineering, K. N. Toosi University of Technology, Tehran, Iran

2 Department of Geodesy and Surveying Engineering, Tafresh University, Tafresh, , Iran


One way to ensure food security is to produce strategic agricultural products on a large scale using industrial methods. Managing large-scale farms consistently and cohesively is a challenging task that requires the utilization of modern technologies. Crop anomalies refer to uncommon and limited factors during agricultural production, leading to localized differentiation in the crop cultivation process. Factors contributing to crop anomalies in agriculture include imbalances in soil nutrients and fertilizers, grazing during crop growth, pests, variations in soil texture and slope in pastures, weed growth, and drought. Detecting and remediating factors limiting crop growth in vast agricultural lands is difficult and these issues are often noticed at harvest time. This article suggests a solution for continuously monitoring of large agricultural fields by analyzing the time series of Sentinel-2 satellite images. The effectiveness of this solution in detecting various anomalies of farms, in agrarian areas has been demonstrated by the results. The proposed solution offers features such as timely diagnosis, the ability to monitor the continuation of irregularities, and the measurement of compensatory measures' effectiveness. The method has successfully identified over five types of anomalies in the selected farms, achieving a detection accuracy of 95.60%.


Main Subjects

Anomalies detection and cause analysis of autumn crops in individual croplands using time series of Sentinel-2 satellite data (Case Study: Golestan province)



Agriculture plays a crucial role in sustaining human livelihood by supplying essential resources such as food, fuel, and raw materials. Given the challenges posed by population growth and limited resources, effective agricultural management becomes imperative in order to fulfill fundamental human requirements. Satellite imagery and remote sensing technology serve as valuable tools for enhancing agricultural product management. These images encompass optical and radar data, which are extensively employed to examine vegetation conditions and, notably, to monitor crop progression. In the realm of agricultural product production, the effective management of factors such as soil, pests, climatic conditions, and unforeseen risks holds significant importance in ensuring sustainable agricultural productivity. In the context of remote sensing images, anomaly detection methods employ unsupervised learning techniques to autonomously identify and uncover uncommon attributes within spectral images. These methods operate in the absence of any prior knowledge about the target scene or spectrum. They rely on the assumption that certain patterns in the input space are more frequently observed, while others occur less frequently. The primary objective of these methods is to uncover and examine these patterns. Anomaly detection techniques can be applied to reflectance, radiance, and other measurement units. Essentially, these methods aim to create a model of the image background and identify pixels that deviate from the background model as anomalies.

Materials and Methods

In this study, the time series data from Sentinel-2 satellite images pertaining to agricultural fields  were utilized. The high temporal resolution offered by the dual multispectral sensors present in the Sentinel-2 satellite platforms (2A and 2B) enabled the generation of a 5-day time series. This research methodology revolves around identifying anomalies in multispectral satellite images and simultaneously gathering field information from farmers through the completion of questionnaires for the selected agricultural lands. This research focuses on two distinct approaches for detecting anomalies. Firstly, individual analysis of images captured at each-time point during the cultivation periods is conducted to identify abnormalities. In the second part, the series of images captured at different stages of the crop year are examined to understand the anomaly formation process. The objective is to assess the effectiveness and sustainability of these two methodologies.

Results and Discussion

Based on the field investigations, agricultural lands experiencing anomalies exhibited poor performance during the corresponding year's harvest. The irregular occurrence of anomalies and the absence of temporal continuity in their detection pose challenges when relying solely on individual time-based images. Based on the cumulative series, anomalies have exhibited continuous presence from the initial detection date throughout the entire observation period, persisting until the end of the agricultural year. Numerous factors have been identified as contributing to the occurrence of farm anomalies. These factors can be categorized into five distinct groups: 1) Anomalies attributed to soil texture and land slope, 2) Anomalies resulting from drought conditions, 3) Anomalies caused by pests and fungi, 4) Anomalies associated with weed infestations, and 5) Multivariate anomalies. Each of these factors exerts its effects, leading to anomalies and their consequences, either persistently or intermittently. Based on the reported findings, anomaly detection methods relying on individual time-based images yielded unfavorable outcomes in both RX and MedSAM detection methods. This challenge arises from the unpredictable occurrence of anomalies throughout a crop year in agricultural fields. However, the utilization of time series-based methods has significantly mitigated this issue. Moreover, the MedSAM method has demonstrated greater success in producing results compared to the RX method.


Utilizing remote sensing technology is a viable and efficient approach to agricultural resource management. anomalies are a common phenomenon observed in agricultural fields, resulting from diverse factors. This article employs the time series data of Sentinel-2 satellite images to investigate the anomalies in autumn crop cultivation at the field level. The results revealed that the utilization of cumulative time series proves to be more effective in detecting and assessing the stability of anomalies compared to methods that analyze individual time points. Examining the temporal pattern of the anomaly detector's response can serve as a potential indicator for assessing the effectiveness of remedial measures in response to anomaly occurrences. The fields chosen for this research predominantly pertain to autumn and dry farming, and the authors of this article have set their research sights on investigating the applicability of this approach to spring crops (which have a more condensed growth period) and water-intensive crops like rice.


Albughdadi, M., Kouamé, D., Rieu, G., & Tourneret, J. Y. (2017, 23-28 July 2017). Missing data reconstruction and anomaly detection in crop development using agronomic indicators derived from multispectral satellite images. Paper presented at the 2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS).
Aslan, M. F., Durdu, A., Sabanci, K., Ropelewska, E., & Gültekin, S. S. (2022). A comprehensive survey of the recent studies with uav for precision agriculture in open fields and greenhouses. Applied Sciences, 12(3), 1047.
Bagheri, N., & Javadi, A. (2022). Assessing the level of technology in the production of strategic agricultural products of IRAN. Industrial Technology Development Quarterly. (inPersian)
Castillo-Villamor, L., Hardy, A., Bunting, P., Llanos-Peralta, W., Zamora, M., Rodriguez, Y., & Gomez-Latorre, D. A. (2021). 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. International journal of applied earth observation and geoinformation, 104, 102535.
Defourny, P., Bontemps, S., Bellemans, N., Cara, C., Dedieu, G., Guzzonato, E., . . . Koetz, B. (2019). Near real-time agriculture monitoring at national scale at parcel resolution: Performance assessment of the Sen2-Agri automated system in various cropping systems around the world. Remote sensing of environment, 221, 551-568. doi:
Gorelick, N., Hancher, M., Dixon, M., Ilyushchenko, S., Thau, D., & Moore, R. (2017). Google Earth Engine: Planetary-scale geospatial analysis for everyone. Remote sensing of environment, 202, 18-27.
Lasaponara, R. (2006). On the use of principal component analysis (PCA) for evaluating interannual vegetation anomalies from SPOT/VEGETATION NDVI temporal series. Ecological modelling, 194(4), 429-434. doi:
León-López, K. M., Mouret, F., Arguello, H., & Tourneret, J.-Y. (2021). Anomaly detection and classification in multispectral time series based on hidden Markov models. IEEE transactions on geoscience and remote sensing, 60, 1-11.
Mir Panahi, S., & Javadi, A. (2022). The need for fundamental changes in Iran's traditional agricultural level in the era of smart agriculture and robotic agriculture. Paper presented at the The 7th International Congress of Agricultural Development and Environment with an emphasis on the development program of nations(inPersian).
Mouret, F., Albughdadi, M., Duthoit, S., Kouamé, D., Poilvé, H., Rieu, G., & Tourneret, J.-Y. (2020). Unsupervised crop anomaly detection at the parcel-level using optical and SAR images: application to wheat and rapeseed crops.
Mouret, F., Albughdadi, M., Duthoit, S., Kouamé, D., Rieu, G., & Tourneret, J.-Y. (2021). Outlier detection at the parcel-level in wheat and rapeseed crops using multispectral and SAR time series. Remote sensing, 13(5), 956.
Racetin, I., & Krtalić, A. (2021). Systematic review of anomaly detection in hyperspectral remote sensing applications. Applied Sciences, 11(11), 4878.
Rodriguez Perez, A. J., Louakfaoui, E. M., Munoz Rastrero, A., Rubio Perez, L. A., & de Pablos Epalza, C. (2004). S.I.I.A. for monitoring crop evolution and anomaly detection in Andalusia by remote sensing (Vol. 5232): SPIE.
San Bautista, A., Fita, D., Franch, B., Castiñeira-Ibáñez, S., Arizo, P., Sánchez-Torres, M. J., . . . Rubio, C. (2022). Crop Monitoring Strategy Based on Remote Sensing Data (Sentinel-2 and Planet), Study Case in a Rice Field after Applying Glycinebetaine. Agronomy, 12(3), 708.
Alter, O., Brown, P. O., & Botstein, D. (2000). Singular value decomposition for genome-wide expression data processing and modeling. Proceedings of the National Academy of Sciences, 97(18), 10101-10106.
Bruckstein, A. M., Donoho, D. L., & Elad, M. (2009). From sparse solutions of systems of equations to sparse modeling of signals and images. SIAM review, 51(1), 34-81.
Chandola, V., Banerjee, A., & Kumar, V. (2009). Anomaly detection: A survey. ACM computing surveys (CSUR), 41(3), 1-58.
Comaniciu, D., & Meer, P. (2002). Mean shift: A robust approach toward feature space analysis. IEEE Transactions on pattern analysis and machine intelligence, 24(5), 603-619.
Ester, M., Kriegel, H.-P., Sander, J., & Xu, X. (1996). A density-based algorithm for discovering clusters in large spatial databases with noise. Paper presented at the kdd.
Filho, J. E. A., Brandão, L. C. P., Fernandes, B. J. T., & Maciel, A. M. A. (2022). A Review of Neural Networks for Anomaly Detection. Ieee Access, 1-1. doi:10.1109/ACCESS.2022.3216007
Ghrib, Z., Jaziri, R., & Romdhane, R. (2020, 19-24 July 2020). Hybrid approach for Anomaly Detection in Time Series Data. Paper presented at the 2020 International Joint Conference on Neural Networks (IJCNN).
Hotelling, H. (1933). Analysis of a complex of statistical variables into principal components. Journal of educational psychology, 24(6), 417.
Japkowicz, N., & Shah, M. (2011). Evaluating Learning Algorithms: A Classification Perspective: Cambridge University Press.
Jia, W., Sun, M., Lian, J., & Hou, S. (2022). Feature dimensionality reduction: a review. Complex & Intelligent Systems, 8(3), 2663-2693.
Kayode Saheed, Y., Harazeem Abdulganiyu, O., & Ait Tchakoucht, T. (2023). A Novel Hybrid Ensemble Learning for Anomaly Detection in Industrial Sensor Networks and SCADA Systems for Smart City Infrastructures. Journal of King Saud University - Computer and Information Sciences. doi:
Matteoli, S., Diani, M., & Corsini, G. (2010). A tutorial overview of anomaly detection in hyperspectral images. IEEE Aerospace and Electronic Systems Magazine, 25(7), 5-28.
Nassif, A. B., Talib, M. A., Nasir, Q., & Dakalbab, F. M. (2021). Machine learning for anomaly detection: A systematic review. Ieee Access, 9, 78658-78700.
Reed, I. S., & Yu, X. (1990). Adaptive multiple-band CFAR detection of an optical pattern with unknown spectral distribution. IEEE transactions on acoustics, speech, and signal processing, 38(10), 1760-1770.
Scholz, M., Kaplan, F., Guy, C. L., Kopka, J., & Selbig, J. (2005). Non-linear PCA: a missing data approach. Bioinformatics, 21(20), 3887-3895.