The extracted points or features are tracked in real time
video by facial feature recognition algorithm of Kanade-LucasTomasi
(KLT) [1, 2, 3]. This algorithm provides the locus of
the variation of recognized features. The algorithm tracks the
most perfect match to the facial detected features provided the
motion of the face is slow. It computes the Eigen values of the
Real time videos from three persons (including the person
whose face image has been stored for comparison) are recorded
for comparison purpose. They are asked to rotate and tilt their
heads in order to observe the effects towards feature
recognition. The number of recognized points starts decaying
with respect to time as the persons have started to rotate their
heads. It has been observed that the number of similar features
those have been paired from the source image decays quickly if
the person in front of the door is not the same person whom,
the facial information and features have been stored. This has
been taken as the key point to separate the person who is the
member of the house and an intruder.