As technology is starting to conquer every strata of the society, the war for protecting confidential data from being intercepted is growing intense by the hour. Biometric- security stands out as the most secure form of authentication in high security zones such as defense, space missions and research head-quarters. Today, forms of password-protection range from face-recognition to retina -scan. Here, we develop a system for recognizing and converting lip movement of an individual into a recognized pattern which is set as a password for the system using image-processing. This system is also a break-through for providing people with motor-disabilities a robust and easy way of protecting their data. By capturing and tracing the successive movement of lips during speech, the corresponding word can be detected. The captured images are represented as points on a two-dimensional flat manifold that enables us to efficiently define the pronunciation of each word and thereby analyze or synthesize the motion of the lips. The motion of lips helps us track the word syllable-by-syllable. With multiple levels of image processing, it becomes possible to set the matching parameters to a very close value, hence not allowing any brute-force or other infamous hacking techniques to break into the user's system. This lip reading technique also serves applications in areas where communication via direct speech is not possible.

Lip-reading lip-contour syllable tracking silent-password threshold analysis image encryption