Region Detection and Tracking

As stated previously, this application was designed with a dual blob tracking in mind. Therefore included in the application is a rough clustering algorithm. All clustering and blob tracking is done using first order and second order moments of centroids. The centroid is calculated to determine velocities and accelerations of blobs as well as the major and minor axis of a constant irradiance ellipse with a tilt.[2]. The centroid is calculated using the standard second order equations:

$$xCenter = \frac {\mu_{10}}{\mu_{00}}, \,\,\, yCenter = \frac {\mu_{01}}{\mu_{00}}$$ (1)

$$Maj = \left( \frac {\mu_{20}+\mu_{02}+[(\mu_{20}-\mu_{02})^2+4\mu_{11}]^{1/2}}{\mu_{00}/2}\right)^{1/2}$$ (2)

$$Min = \left( \frac {\mu_{20}+\mu_{02}-[(\mu_{20}-\mu_{02})^2+4\mu_{11}]^{1/2}}{\mu_{00}/2}\right)^{1/2} \\$$ (3)

$$Tilt = (1/2)\tan^{-1}\left(\frac {2\mu_{11}}{\mu_{20}-\mu_{02}}\right)$$ (4)

To determine the location of the head and hand blob a simple constraint is used. It is known that the right hand (seen on the left of the image), will always be right of the head. Thus, two clusters are computed. The leftmost cluster is determined to be the head blob, while the rightmost is determined to be the hand blob. On the input display, a yellow box denotes the location of the head centroid and the blue box donates the location of the hand centroid. See Figure 4.

  

Figure 4: Head and Gesture Tracking