recognition that will reduce the proxies and also time to mark the attendance.This will also reduce the need to reduce paper pen work for managing documents.
Biometric systems over normal automated system is
preferred to authenticate the user.Real human physical characteristics are almost impossible to change.Passwords, cards are subjected to theft, loss or passing to someone else.
III. OVERVIEW
1) Camera: Camera is used to capture face image of students in which Image containing facewill be used for detection and recognition.
2) Internal Resize: As input will be some fixed size face images. So before using for recognition, images should be resized so that it can be as per the input face image.
3) Face Identification and feature extraction: In face detection, particular face region will be considered and features will be extracted using suitable feature extraction
algorithm.
4) Save template in DB: If input face image will be new then it should be stored in DB. So that further that image will be used in matching process. Automated attendance system using face recognition.
5) Match with template in DB: If face image is already present in DB then for making an attendance it will first match with images in database. And if match found the attendance will be marked.
6) Displayed in Application: After detection and matching if input image found in database then it will be displayed on output screen i.e. in application as an attendance will be marked.
IV. SYSTEM IMPLEMENTATION
A. Algorithms Used :
Here, lattice points whose local patches are inside the image form a set of MRF nodes V. The set of warps Pi can be considered as the set of possible labels for node i. A 4- connected neighbourhood system is then by edges E of the MRF. We also present a method that is classifies whether an input face image is frontal or non-frontal. This method extracts dense SIFT descriptors from the input face image and performs classification using the Support Vector Machine (SVM) algorithm.
Fig1: Neighbouring MRF nodes