In the motive of supporting the blind people we have an idea to propose an audio guidance system in
which the blind people can be [login to view URL] blind people can also able to identify the objects and
face recognition through this [login to view URL] image is captured by the camera which is inbuilt and captured
image is analyzed using Tesseract Optical Character [login to view URL] recognized image is then
converted to voice using open source software Speech Sysnthesizer, eSpeak.
The input will be the queries(who is he? whether known or unknown persons? what object is this?) from
blind people to the smart phone and process(analyze the object ,recognition of face) goes on .Result will
be the voice feedback through earphone to the blind people.
An innovative Facial Normalization algorithm is used to increase the recognition rate of facial
identification and spoofing detection.
Identifying Known and UnKnown [login to view URL] system's database holds a corresponding object of
each images audio [login to view URL] any mismatch or no entry of the image(object and person) in the database
then new image is stored as new dataset with the name specified through microphone inorder to avoid
mismatch of that image(object and person) in the future search.
Incase of known person means it should give the voice output as his name and also should mention in
what state(expression) is he showing to the blind? whether in happy mood or sad or angry like that.
If it is an unknown(new) to the blind ,then that should also able to be identified.
The blind should also able to map. example:map with the help of [login to view URL] navigating if any obstacles
found then it should also intimated through voice output through earphones to the [login to view URL]:
obstacle detection.