Classification of time-series images using deep convolutional neural networks

By Nima Hatami, Yann Gavet, Johan Debayle. International Conference on Machine Vision, 2017

391 citations recorded by the source corpus.

Convolutional Neural Networks (CNN) has achieved a great success in image recognition task by automatically learning a hierarchical feature representation from raw data. While the majority of Time-Series Classification (TSC) literature is focused on 1D signals, this paper uses Recurrence Plots (RP) to transform time-series into 2D texture images and then take advantage of the deep CNN classifier. Image representation of time-series introduces different feature types that are not available for 1D signals, and therefore TSC can be treated as texture image recognition task. CNN model also allows learning different levels of representations together with a classifier, jointly and automatically. Therefore, using RP and CNN in a unified framework is expected to boost the recognition rate of TSC. Experimental results on the UCR time-series classification archive demonstrate competitive accuracy of the proposed approach, compared not only to the existing deep architectures, but also to the state-of-the art TSC algorithms.

Subfields: Artificial Intelligence, Signal Processing

View Paper: https://www.semanticscholar.org/paper/de50861542c40f4180e665b50810a1704c813e49

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Signal Processing

Citations in the corpus

Magnetic Resonance Spectroscopy Quantification using Deep Learning (2018, background): “image representation of spectral data [11]) for improving the accuracy.”
Deep learning for time series classification: a review (2018, methodology): “…extraction method for hand engineering approaches is the transformation of time series into images using specific imaging methods such as Gramian fields (Wang and Oates, 2015b,a), recurrence plots (Hatami et al., 2017; Tripathy and Acharya, 2018) and Markov transition fields (Wang and Oates, 2015).”
Self-Attention for Raw Optical Satellite Time Series Classification (2019, background): “In a temporal context, larger kernel sizes are used where the learned kernel resembles a correlation with learned patterns in the time domain [59, 60] with recent application in remote sensing [54].”
Flight-connection Prediction for Airline Crew Scheduling to Construct Initial Clusters for OR Optimizer (2019, background): “The intuition here is that we can consider each next flight as a different time step, enabling the use of convolutional architecture across time (Hatami et al., 2018; Borovykh et al., 2017).”
Potential, challenges and future directions for deep learning in prognostics and health management applications (2020, background): “Fields (MTF) and Recurrence Plots (RP) have been introduced in [103, 104] as encoding approaches to translate signals to images.”