# Multi-Temporal Land Cover Classification with Sequential Recurrent Encoders

^{*}

## Abstract

**:**

## 1. Introduction

- (i)
- the adaptation of sequence encoders from the field of sequence-to-sequence learning to Earth observation (EO),
- (ii)
- a visualization of internal gate activations on a sequence of satellite observations and,
- (iii)
- the application of crop classification over two seasons.

## 2. Related Work

## 3. Methodology

#### 3.1. Network Architectures and Sequential Encoders

#### 3.2. Prior Work

#### 3.3. This Approach

## 4. Dataset

## 5. Results

#### 5.1. Internal Network Activations

#### 5.2. Quantitative Classification Evaluation

#### 5.3. Qualitative Classification Evaluation

## 6. Discussion

## 7. Conclusions

## Supplementary Materials

## Acknowledgments

## Author Contributions

## Conflicts of Interest

## References

- Zhang, L.; Zhang, Q.; Du, B.; Huang, X.; Tang, Y.Y.; Tao, D. Simultaneous Spectral-Spatial Feature Selection and Extraction for Hyperspectral Images. IEEE Trans. Cybern.
**2018**, 48, 16–28. [Google Scholar] [CrossRef] [PubMed] - Odenweller, J.B.; Johnson, K.I. Crop identification using Landsat temporal-spectral profiles. Remote Sens. Environ.
**1984**, 14, 39–54. [Google Scholar] [CrossRef] - Reed, B.C.; Brown, J.F.; VanderZee, D.; Loveland, T.R.; Merchant, J.W.; Ohlen, D.O. Measuring Phenological Variability from Satellite Imagery. J. Veg. Sci.
**1994**, 5, 703–714. [Google Scholar] [CrossRef] - Bahdanau, D.; Cho, K.; Bengio, Y. Neural Machine Translation by Jointly Learning to Align and Translate. arXiv, 2014; arXiv:1409.0473v7. [Google Scholar]
- Rush, A.; Chopra, S.; Weston, J. A Neural Attention Model for Sentence Summarization. arXiv, 2017; arXiv:1509.00685v2. [Google Scholar]
- Shen, S.; Liu, Z.; Sun, M. Neural Headline Generation with Minimum Risk Training. arXiv, 2016; arXiv:1604.01904v1. [Google Scholar]
- Nallapati, R.; Zhou, B.; dos Santos, C.N.; Gulcehre, C.; Xiang, B. Abstractive Text Summarization Using Sequence-to-Sequence RNNs and Beyond. arXiv, 2016; arXiv:1602.06023v5. [Google Scholar]
- Sutskever, I.; Vinyals, O.; Le, Q.V. Sequence to Sequence Learning with Neural Networks. arXiv, 2014; arXiv:1409.3215v3. [Google Scholar]
- Chorowski, J.; Bahdanau, D.; Serdyuk, D.; Cho, K.; Bengio, Y. Attention-based models for speech recognition. Adv. Neural Inf. Process. Syst.
**2015**, 1, 557–585. [Google Scholar] - Foerster, S.; Kaden, K.; Foerster, M.; Itzerott, S. Crop type mapping using spectral-temporal profiles and phenological information. Comput. Electron. Agric.
**2012**, 89, 30–40. [Google Scholar] [CrossRef] - Conrad, C.; Fritsch, S.; Zeidler, J.; Rücker, G.; Dech, S. Per-Field Irrigated Crop Classification in Arid Central Asia Using SPOT and ASTER Data. Remote Sens.
**2010**, 2, 1035–1056. [Google Scholar] [CrossRef] [Green Version] - Conrad, C.; Dech, S.; Dubovyk, O.; Fritsch, S.; Klein, D.; Löw, F.; Schorcht, G.; Zeidler, J. Derivation of temporal windows for accurate crop discrimination in heterogeneous croplands of Uzbekistan using multitemporal RapidEye images. Comput. Electron. Agric.
**2014**, 103, 63–74. [Google Scholar] [CrossRef] - Hao, P.; Zhan, Y.; Wang, L.; Niu, Z.; Shakir, M. Feature Selection of Time Series MODIS Data for Early Crop Classification Using Random Forest: A Case Study in Kansas, USA. Remote Sens.
**2015**, 7, 5347–5369. [Google Scholar] [CrossRef] - Peña-Barragán, J.M.; Ngugi, M.K.; Plant, R.E.; Six, J. Object-based crop identification using multiple vegetation indices, textural features and crop phenology. Remote Sens. Environ.
**2011**, 115, 1301–1316. [Google Scholar] [CrossRef] - Siachalou, S.; Mallinis, G.; Tsakiri-Strati, M. A hidden markov models approach for crop classification: Linking crop phenology to time series of multi-sensor remote sensing data. Remote Sens.
**2015**, 7, 3633–3650. [Google Scholar] [CrossRef] - Hoberg, T.; Rottensteiner, F.; Feitosa, R.Q.; Heipke, C. Conditional random fields for multitemporal and multiscale classification of optical satellite imagery. IEEE Trans. Geosci. Remote Sens.
**2015**, 53, 659–673. [Google Scholar] [CrossRef] - Zhang, L.; Zhang, L.; Du, B. Deep Learning for Remote Sensing Data: A Technical Tutorial on the State of the Art. IEEE Geosci. Remote Sens. Mag.
**2016**, 4, 22–40. [Google Scholar] [CrossRef] - Zhu, X.X.; Tuia, D.; Mou, L.; Xia, G.S.; Zhang, L.; Xu, F.; Fraundorfer, F. Deep Learning in Remote Sensing: A Comprehensive Review and List of Resources. IEEE Geosci. Remote Sens. Mag.
**2017**, 5, 8–36. [Google Scholar] [CrossRef] - Hu, F.; Xia, G.S.; Hu, J.; Zhang, L. Transferring Deep Convolutional Neural Networks for the Scene Classification of High-Resolution Remote Sensing Imagery. Remote Sens.
**2015**, 7, 14680–14707. [Google Scholar] [CrossRef] - Scott, G.J.; England, M.R.; Starms, W.A.; Marcum, R.A.; Davis, C.H. Training Deep Convolutional Neural Networks for Land-Cover Classification of High-Resolution Imagery. IEEE Geosci. Remote Sens. Lett.
**2017**, 14, 549–553. [Google Scholar] [CrossRef] - Makantasis, K.; Karantzalos, K.; Doulamis, A.; Doulamis, N. Deep Supervised Learning for Hyperspectral Data Classification through Convolutional Neural Networks. In Proceedings of the 2015 IEEE International Geoscience and Remote Sensing Symposium (GARSS), Milan, Italy, 26–31 July 2015; pp. 4959–4962. [Google Scholar]
- Castelluccio, M.; Poggi, G.; Sansone, C.; Verdoliva, L. Land Use Classification in Remote Sensing Images by Convolutional Neural Networks. arXiv, 2015; arXiv:1508.00092. [Google Scholar]
- Lyu, H.; Lu, H.; Mou, L. Learning a Transferable Change Rule from a Recurrent Neural Network for Land Cover Change Detection. Remote Sens.
**2016**, 8, 506. [Google Scholar] [CrossRef] - Jia, X.; Khandelwal, A.; Nayak, G.; Gerber, J.; Carlson, K.; West, P.; Kumar, V. Incremental Dual-memory LSTM in Land Cover Prediction. In Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Halifax, NS, Canada, 13–17 August 2017; pp. 867–876. [Google Scholar]
- Mou, L.; Bruzzone, L.; Zhu, X.X. Learning Spectral-Spatial-Temporal Features via a Recurrent Convolutional Neural Network for Change Detection in Multispectral Imagery. arXiv, 2018; arXiv:1803.02642v1. [Google Scholar]
- Braakmann-Folgmann, A.; Roscher, R.; Wenzel, S.; Uebbing, B.; Kusche, J. Sea Level Anomaly Prediction using Recurrent Neural Networks. arXiv, 2017; arXiv:1710.07099v1. [Google Scholar]
- Sharma, A.; Liu, X.; Yang, X. Land Cover Classification from Multi-temporal, Multi-spectral Remotely Sensed Imagery using Patch-Based Recurrent Neural Networks. arXiv, 2017; arXiv:1708.00813v1. [Google Scholar]
- Rußwurm, M.; Körner, M. Temporal Vegetation Modelling using Long Short-Term Memory Networks for Crop Identification from Medium-Resolution Multi-Spectral Satellite Images. In Proceedings of the IEEE/ISPRS Workshop on Large Scale Computer Vision for Remote Sensing Imagery (EarthVision), Computer Vision and Pattern Recognition Workshops (CVPRW), Honolulu, HI, USA, 21–26 July 2017. [Google Scholar]
- Graves, A.; Wayne, G.; Danihelka, I. Neural Turing Machines. arXiv, 2014; arXiv:1410.5401v2. [Google Scholar]
- Siegelmann, H.; Sontag, E. On the Computational Power of Neural Nets. J. Comput. Syst. Sci.
**1995**, 50, 132–150. [Google Scholar] [CrossRef] - Rafal, J.; Wojciech, Z.; Ilya, S. An Empirical Exploration of Recurrent Network Architectures. In Proceedings of the 32nd International Conference on International Conference on Machine Learning, Lille, France, 6–11 July 2015; Volume 7, pp. 2342–2350. [Google Scholar]
- Hochreiter, S.; Bengio, Y.; Frasconi, P.; Schmidhuber, J. Gradient flow in recurrent nets: The difficulty of learning long-term dependencies. In A Field Guide to Dynamical Recurrent Networks; IEEE Press: New York, NY, USA, 2001; pp. 237–243. [Google Scholar]
- Yoshua, B.; Patrice, S.; Paolo, F. Learning long-term dependencies with gradient descent is difficult. IEEE Trans. Neural Netw.
**1994**, 5, 157–166. [Google Scholar] - Hochreiter, S.; Schmidhuber, J. Long Short-Term Memory. Neural Comput.
**1997**, 9, 1735–1780. [Google Scholar] [CrossRef] [PubMed] - Cho, K.; van Merrienboer, B.; Gulcehre, C.; Bahdanau, D.; Bougares, F.; Schwenk, H.; Bengio, Y. Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation. arXiv, 2014; arXiv:1406.1078v3. [Google Scholar]
- Shi, X.; Chen, Z.; Wang, H.; Yeung, D.Y.; Wong, W.K.; Woo, W.C. Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting. Adv. Neural Inf. Process. Syst.
**2015**, 1, 802–810. [Google Scholar] - Hahnloser, R.; Sarpeshkar, R.; Mahowald, M.A.; Douglas, R.J.; Seung, H.S. Digital selection and analogue amplification coexist in a cortex-inspired silicon circuit. Nature
**2000**, 405, 947–951. [Google Scholar] [CrossRef] [PubMed] - Maas, A.L.; Hannun, A.Y.; Ng, A.Y. Rectifier Nonlinearities Improve Neural Network Acoustic Models. Proc. Int. Conf. Mach. Learn.
**2013**, 28, 6. [Google Scholar] - Kingma, D.P.; Ba, J.L. Adam: A Method for Stochastic Optimization. arXiv, 2014; arXiv:1412.6980v9. [Google Scholar]
- Cohen, J. A coefficient of agreeement for nominal scales. Educ. Psychol. Meas.
**1960**, 20, 37–46. [Google Scholar] [CrossRef] - Karpathy, A.; Johnson, J.; Fei-Fei, L. Visualizing and Understanding Recurrent Networks. arXiv, 2015; arXiv:1506.02078. [Google Scholar]
- Fung, T.; Ledrew, E. The Determination of Optimal Threshold Levels for Change Detection Using Various Accuracy Indices. Photogramm. Eng. Remote Sens.
**1988**, 54, 1449–1454. [Google Scholar] - McHugh, M.L. Interrater reliability: the kappa statistic. Biochem. Med.
**2012**, 22, 276–282. [Google Scholar] [CrossRef] - Ünsalan, C.; Boyer, K.L. Review on Land Use Classification. In Multispectral Satellite Image Understanding: From Land Classification to Building and Road Detection; Springer: London, UK, 2011; pp. 49–64. [Google Scholar]
- Richter, R. A spatially adaptive fast atmospheric correction algorithm. Int. J. Remote Sens.
**1996**, 17, 1201–1214. [Google Scholar] [CrossRef] - Matthew, M.W.; Adler-Golden, S.M.; Berk, A.; Richtsmeier, S.C.; Levine, R.Y.; Bernstein, L.S.; Acharya, P.K.; Anderson, G.P.; Felde, G.W.; Hoke, M.P. Status of Atmospheric Correction using a MODTRAN4-Based Algorithm. In Proceedings of the SPIE Algorithms for Multispectral, Hyperspectral, and Ultra-Spectral Imagery VI, Orlando, FL, USA, 16–20 April 2000; pp. 199–207. [Google Scholar]
- Hagolle, O.; Huc, M.; Villa Pascual, D.; Dedieu, G. A multi-temporal method for cloud detection, applied to FORMOSAT-2, VENuS, LANDSAT and SENTINEL-2 images. Remote Sens. Environ.
**2010**, 114, 1747–1755. [Google Scholar] [CrossRef] [Green Version]

**Figure 1.**Schematic illustration of long short-term memory (LSTM) and gated recurrent unit (GRU) cells analogous to the cell definitions in Table 1. The cell output ${h}_{t}$ is calculated via internal gates and, based on the current input ${x}_{t}$, combined with prior context information ${h}_{t-1}$, ${c}_{t-1}$. This is realized by a concatenation (concat.) of these tensors, as illustrated by merging arrows. LSTM cells are designed to separately accommodate long-term context in the internal cell state ${c}_{t-1}$, from short-term context ${h}_{t-1}$. GRU cells combine all context information in a single, but more sophisticated output ${h}_{t-1}$.

**Figure 2.**Illustrations of recurrent network architectures that inspired this work. The network of previous work [28] shown in (

**a**) creates a prediction ${y}_{t}$ at each observation t based on spectral input information ${x}_{t}$ and the previous context ${h}_{t-1}$, ${c}_{t-1}$. Sequence-to-sequence networks, as shown in (

**b**), aggregate sequential information to an intermediate state ${c}_{T}$, which is a representation of the entire series. (

**a**) Network structure employed in previous work [28]; (

**b**) illustration of a sequence-to-sequence network [8] as often used in neural translation tasks.

**Figure 3.**Schematic illustration of our proposed bidirectional sequential encoder network. The input sequence $x\in \{{x}_{0},\cdots ,{x}_{T}\}$ of observations ${x}_{t}\in {\mathbb{R}}^{h\times w\times d}$ is encoded to a representation ${c}_{T}=[{c}_{T}^{\mathrm{seq}}\parallel {c}_{0}^{\mathrm{inv}}]$. The observations are passed in sequence (seq) and reversed (rev) order to the encoder to eliminate bias towards recent observations. The concatenated representation of both passes ${c}_{T}$ is then projected to softmax-normalized feature maps for each class using a convolutional layer.

**Figure 4.**Area of interest (AOI) north of Munich containing 430 $\mathrm{k}\mathrm{h}\mathrm{a}$ and 137 $\mathrm{k}$field parcels. The AOI is further tiled at multiple scales into datasets for training, validation and evaluation and footprints of individual samples.

**Figure 5.**Information of the area of interest containing location, division schemes, class distributions and dates of acquired satellite imagery. (

**a**) Non-uniform distribution of field classes in the AOI; (

**b**) acquired Sentinel (S2) observations of the twin satellites S2A and S2B.

**Figure 6.**Internal LSTM cell activations of input gate ${i}^{\left(i\right)}$, forget gate ${f}^{\left(i\right)}$, modulation gate ${j}^{\left(i\right)}$ and cell state ${c}^{\left(i\right)}$ at three (of $r=256$) selected cells $i\in \{3,22,47\}$ given the current input ${x}_{t}$ over the sequence of observations $t=\{1,$…$,36\}$. The detail of features at the cell states increased gradually, which indicated the aggregation of information over the sequence. While most cells likely contribute to the classification decision, only some cells are visually interpretable with regard to the current input ${x}_{t}$. One visually-interpretable cell $i=47$ has learned to identify cloud, as input and modulation gates show different activation patterns on cloudy and non-cloudy observations.

**Figure 7.**Confusion matrix of the trained convolutional GRU network on data of the seasons 2016 and 2017. While the confusion of some classes was consistent over both seasons (e.g., winter triticale to winter wheat), other classes are classified at different accuracies for consecutive years (e.g., winter barley to winter spelt).

**Figure 8.**Qualitative results of the convolutional GRU sequential encoder. Examples (

**A**–

**D**) show good classification results. For Example (

**E**) the network misclassified one maize parcel with high confidence, which is indicated by incorrect, but well-defined activations. In a second field, the class activations reveal a confusion between wheat, meadow and maize. For Example (

**F**), most pixels are misclassified. However, the class activations show uncertainty in the classification decision.

**Table 1.**Update formulas of the convolutional variants of standard recurrent neural networks (RNNs), long short-term memory (LSTM) cells and gated recurrent units (GRUs). A convolution between matrices a and b is denoted by $a\ast b$; element-wise multiplication by the Hadamard operator $a\odot b$; and concatenation on the last dimension is marked by $[a\parallel b]$. The activation functions sigmoid $\sigma \left(x\right)$ and tangens hyperbolicus $tanh\left(x\right)$ are used for non-linear scaling.

Gate | Variant | ||
---|---|---|---|

RNN | LSTM [34] | GRU [35] | |

${h}_{t}\leftarrow {x}_{t},{h}_{t-1}$ | ${h}_{t},{c}_{t}\leftarrow {x}_{t},{h}_{t-1},{c}_{t-1}$ | ${h}_{t}\leftarrow {x}_{t},{h}_{t-1}$ | |

Forget/Reset | ${f}_{t}\leftarrow \sigma ([{x}_{t}\parallel {h}_{t-1}]\ast {W}_{f}+1)$ | ${r}_{t}\leftarrow \sigma ([{x}_{t}\parallel {h}_{t-1}]\ast {W}_{r})$ | |

Insert/Update | ${i}_{t}\leftarrow \sigma ([{x}_{t}\parallel {h}_{t-1}]\ast {W}_{i})$ | ${u}_{t}\leftarrow \sigma ([{x}_{t}\parallel {h}_{t-1}]\ast {W}_{u})$ | |

${j}_{t}\leftarrow \sigma ([{x}_{t}\parallel {h}_{t-1}]\ast {W}_{j})$ | |||

Output | ${o}_{t}\leftarrow \sigma ([{x}_{t}\parallel {h}_{t-1}]\ast {W}_{o})$ | ${\tilde{h}}_{t}\leftarrow [{x}_{t}\parallel {r}_{t}\odot {h}_{t-1}]\ast W$ | |

${c}_{t}\leftarrow {c}_{t-1}\odot {f}_{t}+{i}_{t}\odot {j}_{t}$ | |||

${h}_{t}\leftarrow \sigma ([{x}_{t}\parallel {h}_{t-1}]\ast W)$ | ${h}_{t}\leftarrow {o}_{t}\odot tanh({c}_{t})$ | ${h}_{t}\leftarrow {u}_{t}\odot {h}_{t-1}+(1-{u}_{t})\odot tanh({\tilde{h}}_{t})$ |

**Table 2.**Pixel-wise accuracies of the trained convolutional GRU sequential encoder network after training over 60 epochs on data of both growth seasons. The conditional kappa metrics [42] for each class and the overall kappa [40] measure are given for both growth seasons. The best and worst metrics are emphasized by boldface.

Class | Year | |||||||||
---|---|---|---|---|---|---|---|---|---|---|

2016 | 2017 | |||||||||

Precision (User’s Acc.) | Recall (Prod.Acc.) | $\mathit{f}$-Meas. | Kappa | # of Pixels | Precision (User’s Acc.) | Recall (Prod. Acc.) | $\mathit{f}$-Meas. | Kappa | # of Pixels | |

sugar beet | 94.6 | 77.6 | 85.3 | 0.772 | 59 k | 89.2 | 78.5 | 83.5 | 0.779 | 94 k |

oat | 86.1 | 67.8 | 75.8 | 0.675 | 36 k | 63.8 | 62.8 | 63.3 | 0.623 | 38 k |

meadow | 90.8 | 85.7 | 88.2 | 0.845 | 233 k | 88.1 | 85.0 | 86.5 | 0.837 | 242 k |

rapeseed | 95.4 | 90.0 | 92.6 | 0.896 | 125 k | 96.2 | 95.9 | 96.1 | 0.957 | 114k |

hop | 96.4 | 87.5 | 91.7 | 0.873 | 51 k | 92.5 | 74.7 | 82.7 | 0.743 | 53 k |

spelt | 55.1 | 81.1 | 65.6 | 0.807 | 38 k | 75.3 | 46.7 | 57.6 | 0.463 | 31 k |

triticale | 69.4 | 55.7 | 61.8 | 0.549 | 65 k | 62.4 | 57.2 | 59.7 | 0.563 | 64 k |

beans | 92.4 | 87.1 | 89.6 | 0.869 | 27 k | 92.8 | 63.2 | 75.2 | 0.630 | 28 k |

peas | 93.2 | 70.7 | 80.4 | 0.706 | 9 k | 60.9 | 41.5 | 49.3 | 0.414 | 6 k |

potato | 90.9 | 88.2 | 89.5 | 0.876 | 126 k | 95.2 | 73.8 | 83.1 | 0.728 | 140 k |

soybeans | 97.7 | 79.6 | 87.7 | 0.795 | 21 k | 75.9 | 79.9 | 77.8 | 0.798 | 26 k |

asparagus | 89.2 | 78.8 | 83.7 | 0.787 | 20 k | 81.6 | 77.5 | 79.5 | 0.773 | 19 k |

wheat | 87.7 | 93.1 | 90.3 | 0.902 | 806 k | 90.1 | 95.0 | 92.5 | 0.930 | 783 k |

winter barley | 95.2 | 87.3 | 91.0 | 0.861 | 258 k | 92.5 | 92.2 | 92.4 | 0.915 | 255 k |

rye | 85.6 | 47.0 | 60.7 | 0.466 | 43 k | 76.7 | 61.9 | 68.5 | 0.616 | 30 k |

summer barley | 87.5 | 83.4 | 85.4 | 0.830 | 73 k | 77.9 | 88.5 | 82.9 | 0.880 | 91 k |

maize | 91.6 | 96.3 | 93.9 | 0.944 | 919 k | 92.3 | 96.8 | 94.5 | 0.953 | 876 k |

weight.avg | 89.9 | 89.7 | 89.5 | 89.5 | 89.5 | 89.3 | ||||

Overall Accuracy | Overall Kappa | Overall Accuracy | Overall Kappa | |||||||

89.7 | 0.870 | 89.5 | 0.870 |

Approach | Details | |||||
---|---|---|---|---|---|---|

Sensor | Preprocessing | Features | Classifier | Accuracy | # of Classes | |

this work | S2 | none | TOA reflect. | ConvRNN | 90 | 17 |

Ruwurm and Körner [28], 2017 | S2 | atm. cor.(sen2cor) | BOA reflect. | RNN | 74 | 18 |

Siachalou et al. [15], 2015 | LS, RE | geometric correction, image registration | TOA reflect. | HMM | 90 | 6 |

Hao et al. [13], 2015 | MODIS | image reprojection,atm. cor. [45] | statistical phen.features | RF | 89 | 6 |

Conrad et al. [12], 2014 | SPOT, RE, QB | segmentation, atm. cor. [45] | vegetation indices | OBIA + RF | 86 | 9 |

Foerster et al. [10], 2012 | LS | phen. normalization,atm. cor. [45] | NDVI statistics | DT | 73 | 11 |

Pena-Barragán et al. [14], 2011 | ASTER | segmentation,atm. cor. [46] | vegetation indices | OBIA+ DT | 79 | 13 |

Conrad et al. [11], 2010 | SPOT | segmentation,atm. cor. [45] | vegetation indices | OBIA + DT | 80 | 6 |

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Rußwurm, M.; Körner, M.
Multi-Temporal Land Cover Classification with Sequential Recurrent Encoders. *ISPRS Int. J. Geo-Inf.* **2018**, *7*, 129.
https://doi.org/10.3390/ijgi7040129

**AMA Style**

Rußwurm M, Körner M.
Multi-Temporal Land Cover Classification with Sequential Recurrent Encoders. *ISPRS International Journal of Geo-Information*. 2018; 7(4):129.
https://doi.org/10.3390/ijgi7040129

**Chicago/Turabian Style**

Rußwurm, Marc, and Marco Körner.
2018. "Multi-Temporal Land Cover Classification with Sequential Recurrent Encoders" *ISPRS International Journal of Geo-Information* 7, no. 4: 129.
https://doi.org/10.3390/ijgi7040129