1 Introduction
2 Related work
3 Proposed approach
3.1 Preprocessing phase
3.2 Abnormal activity detection phase
Steps of the shape model based on OMEGA equation: | |
---|---|
Step I | The motion region box is designed to include the object of interest and whose axes are aligned with the image axes as shown in Fig. 5d–f. |
Step II | Based on the set of boundary points obtained (i.e., Motion region box), coordinates (C
x
, C
y
) are calculated. |
Step IV | Obtain the distance d = (C
y
− Y
min). |
Step V | Obtain H = half of distance, where H is the window height for extracting the head and shoulder portion of the human object. |
Step VI: | Extract the set of coordinates from the boundary of the upper-segmented contour. |
Step VII | These values are then substituted in Eq. 7 to obtain the pattern for S. |