Univariate Time Series missing data Imputation using Pix2Pix GAN



Time Series, Imputation, cGAN, Rede Pix2Pix


The use of data is essential for the supply of business, scientific and other processes. Often the consumption of these data is hampered when there are sample losses. Aiming to recover values representative of these losses, there are several approaches for filling them. In this paper, we propose a new method for imputation of missing data that transforms time series into an image and thus performs imputation using the conditional generative adversarial network (cGAN) pix2pix GAN. The results of ASMAPE and MAE show that the network outperforms all methods in 50% of the datasets. It was also revealed that the proposed network can learn time series features and retain some advantages over traditional methods, such as imputing the data in its entirety and exploiting spatial and temporal features for imputation.


Download data is not yet available.

Author Biographies

Mauricio Morais Almeida, Universidade Federal do Maranhão

Received a degree in Matematic from theFederal Institute of Education, Science and Technology of Maranhão (IFMA) (2020), He is currently a Master's student in Computer Science at the Federal University of Maranhão conducting research under the guidance of Prof. Dr. João D. S. de Almeida on the issue of imputation of faulty data in time series

João Dallyson Sousa de Almeida, Federal University of Maranhão

Received a degree in Computer Science from the Federal University of Maranhão (UFMA) (2007), a master's degree in Electrical Engineering from UFMA (2010), and a Ph.D. in Electrical Engineering from UFMA (2013). He is currently an Associate Professor I at UFMA. He coordinates the Vision and Image Processing Laboratory (VipLab-UFMA). He has experience in Computer Science, working mainly on the following topics: image processing, machine learning, ophthalmic medical images, and time series.

Geraldo Braz Junior, Federal University of Maranhão

Received an undergraduate degree in Computer Science, a Master's degree in Electrical Engineering with emphasis on Computer Science, and PhD in Electrical Engineering with emphasis on Computer Science, all held at the Federal University of Maranhão (UFMA). He is an Associate Professor I at UFMA, a permanent member of the Post-graduation Programs of Master in Computer Science (PPGCC/UFMA) and Ph.D. in Computer Science / Association UFMA-UFPI. Has experience in Computer Science, working mainly on the following topics: computer vision, machine learning, deep learning, and medical image processing.

Aristofanes Correa Silva, Federal University of Maranhão

Received a bachelor's degree in Computer Science, a master's degree in Electrical Engineering from the Federal University of Maranhão (UFMA), and a PhD in Computer Science from the Pontifical Catholic University of Rio de Janeiro. He is currently a Full Professor at UFMA. He has experience in Computer Science, with emphasis on Graphic Processing (Graphics), working mainly on the following topics: medical imaging and artificial intelligence.

Anselmo Cardoso de Paiva, Federal University of Maranhão

Received a BSc in civil engineering from Maranhão State Univeristy -Brazil in 1990; an MSc in civil engineering-Structures in 1993; and a PhD in Informatics from the Pontiphical Catholic University of Rio de Janeiro – Brazil in 2002. He is currently a Full Professor at the Informatics Department at the Federal University of Maranhão -Brazil. His current interests include medical image processing, geographical information systems and scientific visualization. He is the coordinator of the NCA-UFMA Applied Computing Center. Has experience in Computer Science, with emphasis on Graphics Processing, working mainly on the following topics: Virtual and Augmented Reality, Computer Graphics, GIS, Medical Image Processing and Volumetric Visualization. He is a member of SBC (Brazilian Computer Society) and ACM (Association for Computing Machinery)


E. A. Codling, M. J. Plank, and S. Benhamou, “Random walk models in

biology,” Journal of the Royal society interface, vol. 5, no. 25, pp. 813–

, 2008

R. S. Tsay, “Testing and modeling threshold autoregressive processes,”

Journal of the American statistical association, vol. 84, no. 405, pp. 231–

, 1989.

H. Aboussaid, “The effect of ethnic fractionalization on economic

development. a multilevel analysis,” 2021.

W. A. Fuller, Introduction to statistical time series. John Wiley & Sons,

K. Lakshminarayan, S. A. Harp, R. P. Goldman, T. Samad, et al.,

“Imputation of missing data using machine learning techniques.,” in

KDD, vol. 96, 1996.

S. Rani and A. Solanki, “Data imputation in wireless sensor network

using deep learning techniques,” in Data analytics and management,

pp. 579–594, Springer, 2021.

S. N. Hussain, A. A. Aziz, M. Hossen, N. A. A. Aziz, G. R. Murthy, F. B. Mustakim, et al., “A novel framework based on cnn-lstm neural network for prediction of missing values in electricity consumption time-series datasets.,” Journal of Information Processing Systems, vol. 18, no. 1, 2022.

T. Huamin, D. Qiuqun, and X. Shanzhu, “Reconstruction of time series

with missing value using 2d representation-based denoising autoen-

coder,” Journal of Systems Engineering and Electronics, vol. 31, no. 6,

pp. 1087–1096, 2020.

Z. Guo, Y. Wan, and H. Ye, “A data imputation method for multivariate

time series based on generative adversarial network,” Neurocomputing,

vol. 360, pp. 185–197, 2019.

S. Kawakura and R. Shibasaki, “Deep learning-based self-driving car:

Jetbot with nvidia ai board to deliver items at agricultural workplace

with object-finding and avoidance functions,” European Journal of

Agriculture and Food Sciences, vol. 2, no. 3, 2020.

Y. Kortli, M. Jridi, A. Al Falou, and M. Atri, “Face recognition systems:

A survey,” Sensors, vol. 20, no. 2, p. 342, 2020.

Z. Wang and T. Oates, “Imaging time-series to improve classification

and imputation,” in Twenty-Fourth International Joint Conference on

Artificial Intelligence, 2015.

N. Hatami, Y. Gavet, and J. Debayle, “Classification of time-series

images using deep convolutional neural networks,” in Tenth international

conference on machine vision (ICMV 2017), vol. 10696, p. 106960Y,

International Society for Optics and Photonics, 2018.

Y. Zhuang, R. Ke, and Y. Wang, “An innovative method for traffic

data imputation based on convolutional neural network,” IET Intelligent

Transport Systems, vol. 13, 11 2018.

Y. Luo, X. Cai, Y. Zhang, J. Xu, et al., “Multivariate time series

imputation with generative adversarial networks,” Advances in neural

information processing systems, vol. 31, 2018.

E. Brophy, Z. Wang, and T. E. Ward, “Quick and easy time se-

ries generation with established image-based gans,” arXiv preprint

arXiv:1902.05624, 2019.

E. Brophy, Z. Wang, Q. She, and T. Ward, “Generative adversarial

networks in time series: A survey and taxonomy,” arXiv preprint

arXiv:2107.11098, 2021.

S. Oehmcke, T.-H. K. Chen, A. V. Prishchepov, and F. Gieseke,

“Creating cloud-free satellite imagery from image time series with deep

learning,” in Proceedings of the 9th ACM SIGSPATIAL International

Workshop on Analytics for Big Geospatial Data, pp. 1–10, 2020.

R. C. Pereira, M. S. Santos, P. P. Rodrigues, and P. H. Abreu, “Reviewing

autoencoders for missing data imputation: Technical trends, applications

and outcomes,” Journal of Artificial Intelligence Research, vol. 69,

pp. 1255–1285, 2020.

Y. Xia, L. Zhang, N. Ravikumar, R. Attar, S. K. Piechnik, S. Neubauer,

S. E. Petersen, and A. F. Frangi, “Recovering from missing data

in population imaging–cardiac mr image imputation via conditional

generative adversarial nets,” Medical Image Analysis, vol. 67, p. 101812,

A. T. Elergone, “ElectricityLoadDiagrams20112014 Data Set,” 2020.

K. Dunn, “OpenMV.net Datasets.” https://openmv.net/, 2018.

H. Fanaee-T and J. Gama, “Event labeling combining ensemble detectors

and background knowledge,” Progress in Artificial Intelligence, pp. 1–

, 2013.

D. H. Stolfi, E. Alba, and X. Yao, “Predicting car park occupancy rates

in smart cities,” in International Conference on Smart Cities, pp. 107–

, Springer, 2017.

KANKANA, “Daily minimum temperatures in me.”

J. Brownlee,“Machine learning datasets.”


sunspots.csv, dez 2020. accessed: 20-12-2021.

S. Moritz, M. Friese, A. Fischbach, C. Schlitt, and T. Bartz-Beielstein,

“GECCO Industrial Challenge 2015 Dataset: A heating system dataset

for the ’Recovering missing information in heating system operating

data’ competition at the Genetic and Evolutionary Computation Confer-

ence 2015, Madrid, Spain,” May 2015.

P. Isola, J.-Y. Zhu, T. Zhou, and A. A. Efros, “Image-to-image translation

with conditional adversarial networks,” in Proceedings of the IEEE

conference on computer vision and pattern recognition, pp. 1125–1134,

M. Mirza and S. Osindero, “Conditional generative adversarial nets,”

arXiv preprint arXiv:1411.1784, 2014.

A. Flores, H. Tito, and C. Silva, “Local average of nearest neighbors:

Univariate time series imputation,” International Journal of Advanced

Computer Science and Applications, vol. 10, no. 8, pp. 45–50, 2019.

C. J. Willmott and K. Matsuura, “Advantages of the mean absolute error

(mae) over the root mean square error (rmse) in assessing average model

performance,” Climate research, vol. 30, no. 1, pp. 79–82, 2005.

V. Kreinovich, H. T. Nguyen, and R. Ouncharoen, “How to estimate

forecasting quality: A system-motivated derivation of symmetric mean

absolute percentage error (smape) and other similar characteristics,”

M. Daraghmeh, A. Agarwal, R. Manzano, and M. Zaman, “Time series

forecasting using facebook prophet for cloud resource management,”

in 2021 IEEE International Conference on Communications Workshops

(ICC Workshops), pp. 1–6, IEEE, 2021.

S. Zhang, L. Gong, Q. Zeng, W. Li, F. Xiao, and J. Lei, “Imputation of

gps coordinate time series using missforest,” Remote Sensing, vol. 13,

no. 12, p. 2312, 2021.



How to Cite

Morais Almeida, M., Sousa de Almeida, J. D., Braz Junior, G., Correa Silva, A., & Cardoso de Paiva, A. (2023). Univariate Time Series missing data Imputation using Pix2Pix GAN. IEEE Latin America Transactions, 21(3), 505–512. Retrieved from https://latamt.ieeer9.org/index.php/transactions/article/view/7152