Face Recognition in the Scrambled Domain Using MK-RDA and ANN

Kavita Kadam, C.R Jadhav

Abstract


Facial look identity is a vital mission by means of human-interacting structures that goal to be aware of versions within the human’s emotional state. the principle challenge or the crucial part in surveillance society is the privacy-shielding era. because the rapid improvement in the internet international it turns into very essential to scramble the pics in the video or files for the duration of transmission. in this the biometric identity of photographs or faces from scrambled pictures plays a completely tough mission. Numbers of various technology are carried out to provide privateness for the duration of surveillance or during transmission of video however they're lack of essential traits, like reversibility or visible fine maintenance. in lots of scrambling methods the faces are covered by a few animation which may additionally or may not cover all faces or it receives hard to recover pics from this technique. Many guide method also are us used by which we will unscramble an photo but they are no longer powerful that a good deal. to overcome all this matters we proposed a novel approach- Many-Kernel Random Discriminate analysis (MK-RDA) to find out discriminative patterns from chaotic indicators. structures get better accuracy bring about best photos. To PIE and ORL datasets has getting above ninety% accuracy.

Full Text:

PDF


DOI: http://doi.org/10.11591/ijict.v7i1.pp16-23

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

View IJICT Stats