The Use of Multi-temporal Spectral Information to Improve the Classification of Agricultural Crops in Landscapes
- 49 Downloads
Machine learning opens up a wide range of possibilities for crop classification and mapping using satellite data. With the shortening of their revisit cycles, satellites are now able to provide an increasing amount of data with valuable temporal information. We propose a machine learning approach to efficiently analyze multi-temporal data for crop identification and monitoring. This methodology utilizes a Bayesian approach to gradually improve classification accuracy as the temporal resolution increases. Two multispectral satellite configurations were simulated with hyperspectral data and analyzed with a support vector machine approach and a deep learning algorithm. Results showed that both approaches are able to efficiently process information as time progresses and rapidly achieve very high accuracies. The deep learning algorithm has the advantage that the dynamic component, time, is accounted for automatically, without the need of being actively incorporated by the analyst.
KeywordsHyperspectral data Temporal data Machine learning Support Vector Machine Deep learning LSTM Remote sensing
This work was supported by the Federal Ministry of Food (BMELV) and Agriculture and the Ministry of Science, Research and Culture (MWKF) of the State of Brandenburg. Furthermore, I would like to thank our former colleague of the ZALF: Bernd Zbell, who did the data sampling and the preparation of the Excel-tables. My special thanks to the Python community which developed the used software and made it as Free and Open Source Software available.
- Abadi M, Agarwal A, Barham P, Brevdo E, Chen Z, Citro C, Corrado GS, Davis A, Dean J, Devin M, Ghemawat S, Goodfellow IJ, Harp A, Irving G, Isard M, Jia Y, Jozefowicz R, Kaiser L, Kudlur M, Levenberg J, Mané D, Monga R, Moore S, Murray DG, Olah C, Schuster M, Shlens J, Steiner B, Sutskever Il, Talwar K, Tucker PA, Vanhoucke V, Vasudevan V, Viégas F, Vinyals O, Warden P, Wattenberg M, Wicke M, Yu Y, Zheng X (2016) TensorFlow: large-scale machine learning on heterogeneous distributed systems. CoRR, Volume abs/1603.04467. http://arxiv.org/abs/1603.04467
- Becerra-García RA, García-Bermúdez RV, Joya-Caparrós G, Fernández-Higuera A, Velázquez-Rodríguez C, Velázquez-Mariño M, Cuevas-Beltrán FR, García-Lagos F, Rodráguez-Labrada R (2017) Data mining process for identification of non-spontaneous saccadic movements in clinical electrooculography. Neurocomputing 250:28–36Google Scholar
- Chen T, Guestrin C (2016) XGBoost: a scalable tree boosting system. In: KDD 16 proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining, San Francisco, California, USA, August 13–17, 2016. ACM DL, Pages 785–794Google Scholar
- Collet F, others (2015) Keras. https://keras.io
- Gilbertson J (2017) Machine learning for object-based crop classification using multi-temporal Landsat-8 imagery. MSc. thesis, Stellenbosch University, 102 ppGoogle Scholar
- Griffiths P, Nendel C, Hostert P (2018) Intra-annual reflectance composites from Sentinel-2 and Landsat for national-scale crop and land cover mapping. Remote Sens Environ, January 2019:135–151. (Open access)Google Scholar
- Hütt C, Waldhoff G (2018) Multi-data approach for crop classification using multitemporal, dual-polarimetric TerraSAR-X data, and official geodata. Eur J Remote Sens 51:62–74Google Scholar
- Ji S, Zhang C, Xu A, Duan Y (2018) 3D convolutional neural networks for crop classification with multi-temporal remote sensing images. Remote Sens 10(75):1–17Google Scholar
- Mura M, Bottalico F, Giannetti F, Bertani R, Giannini R, Mancini M, Orlandini S, Travaglini D, Chirici G (2018) Exploiting the capabilities of the Sentinel-2 multi spectral instrument for predicting growing stock volume in forest ecosystems. Int J Appl Earth Obs Geoinf 66:126–134CrossRefGoogle Scholar
- Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, Blondel M, Prettenhofer P, Weiss R, Dubourg V, Vanderplas J, Passos A, Cournapeau D, Brucher M, Perrot M, Duchesnay E (2011) Scikit-learn: machine learning in Python. J Mach Learn Res 12:2825–2830Google Scholar
- Sharma A, Liu X, Yang X (2018) Land cover classification from multi-temporal, multi-spectral remotely sensed imagery using patch-based recurrent neural networks. Neural Netw 105:346–355Google Scholar