A Simple Efficient Approximation Algorithm for Dynamic Time Warping

被引:7
|
作者
Ying, Rex [1 ]
Pan, Jiangwei [2 ]
Fox, Kyle [2 ]
Agarwal, Pankaj K. [2 ]
机构
[1] Stanford Univ, Stanford, CA 94305 USA
[2] Duke Univ, Durham, NC 27706 USA
来源
24TH ACM SIGSPATIAL INTERNATIONAL CONFERENCE ON ADVANCES IN GEOGRAPHIC INFORMATION SYSTEMS (ACM SIGSPATIAL GIS 2016) | 2016年
关键词
Curve matching; dynamic time warping; approximation algorithm; trajectory analysis; SETS;
D O I
10.1145/2996913.2996954
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Dynamic time warping (DTW) is a widely used curve similarity measure. We present a simple and efficient (1 + is an element of)approximation algorithm for DTW between a pair of point sequences, say, P and Q, each of which is sampled from a curve. We prove that the running time of the algorithm is O (k(2) /is an element of n log sigma) for a pair of k-packed curves with a total of n points, assuming that the spreads of P and Q are bounded by sigma. The spread of a point set is the ratio of the maximum to the minimum pairwise distance, and a curve is called k-packed if the length of its intersection with any disk of radius r is at most kr. Although an algorithm with similar asymptotic time complexity was presented in [1], our algorithm is considerably simpler and more e ffi cient in practice. We have implemented our algorithm. Our experiments on both synthetic and real_world data sets show that it is an order of magnitude faster than the standard exact DP algorithm on point sequences of length 5,000 or more while keeping the approximation error within 5-10%. We demonstrate the e ffi cacy of our algorithm by using it in two applications | computing the k most similar trajectories to a query trajectory, and running the iterative closest point method for a pair of trajectories. We show that we can achieve 8-12 times speedup using our algorithm as a subroutine in these applications, without compromising much in accuracy.
引用
收藏
页数:10
相关论文
共 50 条
  • [21] A Dynamic Time Warping Algorithm for Industrial Robot Motion Analysis
    Johnen, Benjamin
    Kuhlenkoetter, Bernd
    2016 ANNUAL CONFERENCE ON INFORMATION SCIENCE AND SYSTEMS (CISS), 2016,
  • [22] Self-supervised dynamic time-warping algorithm
    Zhu, X.-Y.
    Cahill, L.W.
    Journal of Electrical and Electronics Engineering, Australia, 1992, 12 (03): : 300 - 307
  • [23] Dynamic Time Warping: A novel algorithm for rhythm discrimination in ICDs
    Kamousi, Baharan
    Lin, Bryant
    Lee, Gregory
    Ai-Ahmad, Amin
    Hsia, Henry H.
    Zei, Paul C.
    Wang, Paul J.
    JOURNAL OF THE AMERICAN COLLEGE OF CARDIOLOGY, 2008, 51 (10) : A14 - A14
  • [24] MORPHOLOGICAL MEASUREMENT OF THE SEP USING A DYNAMIC TIME WARPING ALGORITHM
    EISEN, A
    ROBERTS, K
    LAWRENCE, P
    ELECTROENCEPHALOGRAPHY AND CLINICAL NEUROPHYSIOLOGY, 1986, 65 (02): : 136 - 141
  • [25] A robust Dynamic Time Warping algorithm for batch trajectory synchronization
    Zhang, Yang
    Edgar, Thomas F.
    2008 AMERICAN CONTROL CONFERENCE, VOLS 1-12, 2008, : 2864 - 2869
  • [26] ON MINIMIZING MEMORY IN SYSTOLIC ARRAYS FOR THE DYNAMIC TIME WARPING ALGORITHM
    DIJKSTRA, E
    PIGUET, C
    INTEGRATION-THE VLSI JOURNAL, 1986, 4 (02) : 155 - 173
  • [27] Application of dynamic time warping algorithm for pattern similarity of gait
    Lee, Hyun-Seob
    JOURNAL OF EXERCISE REHABILITATION, 2019, 15 (04) : 526 - 530
  • [28] An OGS-based Dynamic Time Warping Algorithm for Time Series Data
    Zhou, Mi
    2013 INTERNATIONAL CONFERENCE ON ENGINEERING, MANAGEMENT SCIENCE AND INNOVATION (ICEMSI 2013), 2013,
  • [29] Efficient Subsequence Join Over Time Series Under Dynamic Time Warping
    Vo Duc Vinh
    Duong Tuan Anh
    RECENT DEVELOPMENTS IN INTELLIGENT INFORMATION AND DATABASE SYSTEMS, 2016, 642 : 41 - 52
  • [30] DYNAMIC POSITIONAL WARPING: DYNAMIC TIME WARPING FOR ONLINE HANDWRITING
    Chang, Won-Du
    Shin, Jungpil
    INTERNATIONAL JOURNAL OF PATTERN RECOGNITION AND ARTIFICIAL INTELLIGENCE, 2009, 23 (05) : 967 - 986