University Affiliation Assignment PDF
15 Pages3865 Words113 Views
Added on 2021-06-15
University Affiliation Assignment PDF
Added on 2021-06-15
ShareRelated Documents
UNIVERSITY AFFILIATION2018 CSE3/4 VIS VISUAL INFORMATION SYSTEMS ASSIGNMENTSTUDENT NAME
TABLE OF CONTENTSBASIC OVERVIEW.......................................................................................................................2Eigenface techniques for face recognition...................................................................................2RESULTS AND OBSERVATIONS: MATLAB IMPLEMENTATION.......................................3PROCEDURE:.............................................................................................................................3Experiment and results.................................................................................................................4Part 1: Resizing the images stored in the training dataset and test dataset..................................4Part 2: Determining K1 and K2...................................................................................................6Part 3: Training and Test dataset results from classifier output...................................................7DISCUSSION..................................................................................................................................7System performance.....................................................................................................................7REFERENCES................................................................................................................................91
BASIC OVERVIEWEigenface techniques for face recognitionThe modern security systems incorporate the use of facial recognition as a way of enablingauthorization. This can be achieved in personal identification for instance, at the immigrationoffice or border points, for human computer interactions such as phone unlock using facialrecognition, and major security systems for instance, in offices or large residential complex orpenthouses[ CITATION Nav02 \l 1033 ]. Every human being has very complex, multidimensional,and meaningful facial unique attributes that differentiate them from others. This makes the facialrecognition process more difficult to implement in security systems. Some of the local featuresthat the recognition process focusses on are the eyes, nose, and mouth before extracting thefeature of the whole face[ CITATION Cha10 \l 1033 ]. There are a number of approaches defined to enable the facial recognition by systems. Forinstance, using artificial intelligence, one can create neural networks and self-organizing maps,SOMs, where the user trains some images and uses test images to test the recognition. Otheralternatives to this approach are content-based image retrieval, principal component analysis andthe relevance feedback[ CITATION Tur1 \l 1033 ]. There are three stages performed in the process offacial recognition such as the face location detection, feature extraction, and facial imageclassification. The face recognition is done using eigenface algorithm[ CITATION Ema \l 1033 ].The face images are projected into a feature space that best encodes the variation among theknown face images. The face space is well expounded by the eigenfaces, which are theeigenvectors of the set of faces. The eigenface algorithm computes the average face, v. Thealgorithm collects the difference between training images and the average face[ CITATION Rui02 \l1033 ]. The differences are saved in a matrix where M is the number of pixels and N is thenumber of stored or trained images. The algorithm for eigenfaces is denoted by the equation,A=[u11−v,...,un1−v,...,u1p−v,...,unp−v]The eigenvectors of the covariance matrix C are used to give the final eigenfaces. This is doneusing powerful tools with a stable runtime such as MATLAB R2017a. Therefore, C=AAT2
There are N-1 meaningful eigenvectors, when the number of data points is smaller than thedimensions. To get a faster response on the value of the eigenvectors of C,L=ATAThe training face images and the new face images are represented as a linear combination of theeigenfaces. For instance, a face image, u, can be represented as,u=∑iaiφiThe eigenvectors are usually orthogonal to the eigenvalues such that,ai=uTφiThe PCA seeks directions that are efficient for the representation of the data and seeks tomaximize the total scatter. The PCA reduces the dimension of the data and speeds up thecomputational time. The time taken to perform facial recognition is important especially inimplementation in the actual environment such that the systems should use the least time todetect a face[ CITATION Tol06 \l 1033 ]. It should be close to real-time. RESULTS AND OBSERVATIONS: MATLAB IMPLEMENTATIONThe task aims at using the training and testing data to identify and extract face images from theimage saved. This is done when the algorithm pulls similar images from the database with a setof 30 images from the training data and 20 images from the test data. PROCEDURE:(i)The training set is obtained as the 30 images and the test set has 20 images all definedin set folders. The algorithm is loaded and the eigenfaces are calculated using thePCA projections. These projections define the eigenspace. (ii)The new face is checked by using the test data and the weight of the connections orlinks is computed. The system, at this point, determines if the image used is a face.When the algorithm identifies the new image as a face, the weight pattern is groupedunder the known or unknown[ CITATION Hon05 \l 1033 ]. 3
End of preview
Want to access all the pages? Upload your documents or become a member.
Related Documents
Lamarckian Genetic Algorithm in Neural Networking for Face Detection Systemlg...
|14
|3213
|305
Cyber Security and Analytics Assignment 2022lg...
|4
|1195
|13