This paper is concerned with detecting the period of cyclic object motion in a short video or sequence with a limited number of frames. This problem can be studied with either frequency-domain methods or time-domain methods. A frequency-domain method is fundamentally limited in terms of frequency resolution—especially with a small number of frames—and its ability to handle a periodic impulsive or spiky signal. Existing time-domain methods are primarily based on an analysis of the autocorrelation function of a signal and can be sensitive to noise in the signal. In this paper, we offer an alternative time-domain method. Rather than using autocorrelation as the basis, our proposed method uses peak analysis. Specifically, after computing the similarity between a reference image and those in the sequence, our algorithm applies one of two period detection procedures—one based on clustering and the other on watershed to analyze the peaks of the similarity time series—in estimating the period of object motion embedded in the similarity function. Video sequences from three different applications are used to establish the feasibility of our proposed algorithm and its superiority to competing algorithms.
[1]
G. D. Bergland,et al.
A guided tour of the fast Fourier transform
,
1969,
IEEE Spectrum.
[2]
David Gerhard,et al.
Pitch Extraction and Fundamental Frequency: History and Current Techniques
,
2003
.
[3]
M. Sondhi,et al.
New methods of pitch extraction
,
1968
.
[4]
Paul A. Viola,et al.
Alignment by Maximization of Mutual Information
,
1997,
International Journal of Computer Vision.
[5]
M. Ross,et al.
Average magnitude difference function pitch extractor
,
1974
.
[6]
Mubarak Shah,et al.
Cyclic motion detection for motion based recognition
,
1994,
Pattern Recognit..
[7]
Hajime Kobayashi,et al.
Weighted autocorrelation for pitch extraction of noisy speech
,
2001,
IEEE Trans. Speech Audio Process..
[8]
Ronald W. Schafer,et al.
Real-time digital hardware pitch detector
,
1976
.