{"references": ["Sakoe, H. and Chiba, S. (1978) Dynamic Programming Algorithm Optimization for Spoken Word Recognition, DOI: 10.1109/TASSP.1978.1163055", "Quenot, G. M. (1992) The \"Orthogonal Algorithm\" for Pptical Flow Detection using Dynamic Programming, ISBN: 0780305329", "Quenot, G. M. (1996) Computation of Optical Flow using Dynamic Programming, IAPR Workshoop on Machine Vision Applications", "Quenot, G. M., Pakleza, J. and Kowalewski, T. A. (1998) Particle image velocimetry with optical flow, DOI: 10.1007/s003480050222", "Quenot, G (2000) Synchronous Orthogonal Dynamic Programming for Particle Image Velocimetry, 9th International Symposium on Flow Visualisation", "Kriete, D. M. et al (2018) Extracting the turbulent flow-field from beam emission spectroscopy images using velocimetry, DOI: 10.1063/1.5036535"]}

The authors wish to thank: Dr Edward Higgins for helping with the initial `f2py` port of some of the routine's functions; Yorick Enters and Matthew Hill for their work on initial testing of the routine and providing feedback that lead to improvements in the library; and Dr. Peter Hill of the PlasmaFair project for assisting in making this library widely available for use by other researchers.

DTW is a dynamic time warping library designed for tokamak velocimetry measurements. The algorithm works by distorting image 2 into image 1, then using the distorted coordinates to compute the displacements required to distort image 1 into image 2, given at the positions in image 1.