/.Amd mnt/lotus/host/home/jaishakthi/presentation/rmeet2/rmeet2

658 views

Published on

My second

Published in: Technology, News & Politics
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
658
On SlideShare
0
From Embeds
0
Number of Embeds
3
Actions
Shares
0
Downloads
0
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

/.Amd mnt/lotus/host/home/jaishakthi/presentation/rmeet2/rmeet2

  1. 1. A Novel Approach Using PCA And SVM for Face Detection A Novel Approach Using PCA And SVM for Face Detection S.M. Jaisakthi September 9, 2009
  2. 2. A Novel Approach Using PCA And SVM for Face Detection Introduction Intoduction Face Detection is Pattern recognition problem Applicable in Bankcard Identification System,Security Monitoring,Computer Vision etc. Difficult Problem Broadly classified as Appearence Based Approach Feature Based Approach Moment Based Approach
  3. 3. A Novel Approach Using PCA And SVM for Face Detection Algorithm Algorithm In the proposed algorithm 1 Identifies the face potential area 2 Calculates eigenvector using PCA 3 Obtained eigenvector is trained with SVM
  4. 4. A Novel Approach Using PCA And SVM for Face Detection Algorithm Face Potential Area Face Potential Area Pixel character is different for face and non-face image scan the testimage using sliding window and crop the image of specified size calculate histogram distribution for each subimage face and non-face area has different histogram distribution face is then cropped
  5. 5. A Novel Approach Using PCA And SVM for Face Detection Algorithm Face Potential Area Histogram Distribution
  6. 6. A Novel Approach Using PCA And SVM for Face Detection Principal Component Analysis Principal Component Analysis(PCA) Common techinque for finding patterns Compresses a set of high dimensional vectors into a set of lower dimensional vectors Computing PCA Organize the data set Calculate the empirical mean Calculate the deviations from the mean Find the covariance matrix Find the eigenvectors and eigenvalues of the covariance matrix Sort the eigenvalues and the corresponding eigenvectors Select first d≤n eigenvectors The projected test image is compared to every projected training image by using similarity measure
  7. 7. A Novel Approach Using PCA And SVM for Face Detection Principal Component Analysis Support Vector Machine Finds optimal hyperplane that best separates two class Find support vector inorder to find optimal hyperplane Non-linear case, kernal functions are used
  8. 8. A Novel Approach Using PCA And SVM for Face Detection Principal Component Analysis Face detection 1 Face potential area is selected 2 PCA is used to decrease the dimension of face feature space. 3 SVM is used as classifier
  9. 9. A Novel Approach Using PCA And SVM for Face Detection Principal Component Analysis Conclusion Results in high performance saves detection time
  10. 10. A Novel Approach Using PCA And SVM for Face Detection Principal Component Analysis THANK YOU

×