10-08-2015, 09:36 PM
RESPECTED SIR,
I WANT PDF TYPE DOCUMENT ABOUT FACE DETECTING ATTENDANCE SOFTWARE PROJECT TO WRITE 'SRS' FOR THIS PROJECT
|
face recognition in e attendance srs
|
RESPECTED SIR,
I WANT PDF TYPE DOCUMENT ABOUT FACE DETECTING ATTENDANCE SOFTWARE PROJECT TO WRITE 'SRS' FOR THIS PROJECT
23-03-2017, 04:15 PM
It is difficult to estimate the assistance precisely using each face recognition result independently because the face detection rate is not high enough. In this article we propose a method to estimate the attendance precisely using all the results of the facial recognition obtained by continuous observation. Continuous observation improves performance for estimating attendance. We build the class attendance system based on facial recognition and apply the system to the class. In this paper, the related work in the field of assistance management and facial recognition are first reviewed. Then introduce our structure and system plan. Finally, the experiments are implemented as evidence to support our plan. The result shows that continuous observation improved performance for attendance estimation.
Recently, facial recognition is attracting a lot of attention in society access to multimedia network information. As network security, content indexing and video recovery and compression benefits from face recognition technology because people are the focus of many video. Control through facial recognition not only makes hackers virtually impossible to steal your "password", but also increases ease of use in human-computer interaction. A training set consists of facial images from which a set of eigen faces can be generated by performing a mathematical process called principal component analysis (PCA) in a large set of images representing different human faces. New faces can project a new image (half subtracted) on the eigenfaces and thus record how that new face differs from the middle face
The eigenvalues associated with each eigenface represent how much the images in the training set vary from the average image in that direction. We lose information by projecting the image into a subset of the eigenvectors, but we minimize this loss by keeping those eigen faces with the highest eigenvalues. In this way, the input image of each student corresponds to the image of the training set. If a match is found, the attendance is marked as present, but marked as absent for that particular student. This overcomes power, identifies punctuality of students and also recognizes teachers and their punctuality and regularity in the management of classes. Administration can know about handling classes in your chair
|
« Next Oldest | Next Newest »
|
