Real-time Visual Measurement: Target Counting / Rotation Measurement
Summary
Real-time observation of multiple targets in image sequences at high frame rates is possible with our multi-target tracking using a vision chip. Two applications based on this image processing are described in this page.
Target Counting
This Target counting is a measurement that counts the number of regions that newly appear in an image. This counting is realized at high-frame rate by using the multi-target tracking.
The tracking process in target counting ensures the accuracy of the measurement. In addition, image processing on a vision chip enables the counting of targets with desired shape and size. For example, as an application of this target counting we consider the counting of small fish swimming over a line. This kind of application is very difficult to automate because the targets move very quickly and irregularly.
Fig. 1: Target counting system for experiment and its results
Rotation measurement
This rotation measurement obtains rotation information on a rotating object in real-time. The rotation information is the rotation axis and rotation speed.
This measurement estimates the rotation information from the distribution of speed vectors on a sphere. The vectors are obtained by multi-target tracking.
As an application of this rotation measurement, we consider ball games. For example, the rotation speed of a soccer ball in a free kick reaches about 600 rpm and a baseball at pitching reaches about 1800 rpm. The experiment results show that our rotation measurement is accurate up to about 1200 rpm.
Fig. 2: Rotation measurement system for experiment and its results
Movies
References
- Yoshihiro Watanabe, Takashi Komuro, Shingo Kagami, Masatoshi Ishikawa: Multi-Target Tracking Using a Vision Chip and its Applications to Real-Time Visual Measurement, Journal of Robotics and Mechatronics, Vol. 17, No. 2 , pp. 121-129, 2005. [PDF]
- Yoshihiro Watanabe, Takashi Komuro, Shingo Kagami, Masatoshi Ishikawa: Real-time Visual Measurements using High-speed Vision, Optics East (Philadelphia 2004.10.28) / Machine Vision and its Optomechatronic Applications, Proceedings of SPIE Vol. 5603, pp. 234-242. [PDF]
