AUTOMATIC HUMAN FACE RECOGNITION USING MULTIVARIATE GAUSSIAN MODEL AND FISHER LINEAR DISCRIMINATIVE ANALYSIS
Journal: ICTACT Journal on Image and Video Processing (IJIVP) (Vol.5, No. 1)Publication Date: 2014-08-01
Authors : Surya Kasturi; Ashoka Vanjare; S. N. Omkar;
Page : 899-902
Keywords : Face Recognition; Multivariate Gaussian Model; Image Segmentation; Connected Component; Color Spaces;
Abstract
Face recognition plays an important role in surveillance, biometrics and is a popular application of computer vision. In this paper, color based skin segmentation is proposed to detect faces and is matched with faces from the dataset. The proposed color based segmentation method is tested in different color spaces to identify suitable color space for identification of faces. Based on the sample skin distribution a Multivariate Gaussian Model is fitted to identify skin regions from which face regions are detected using connected components. The detected face is match with a template and verified. The proposed method Multivariate Gaussian Model ? Fisher Linear Discriminative Analysis (MGM ? FLDA) is compared with machine learning - Viola & Jones algorithm and it gives better results in terms of time.
Other Latest Articles
- TWO STAGE SPATIAL DOMAIN IMAGE FUSION TECHNIQUES
- VLSI ARCHITECTURE FOR IMAGE COMPRESSION THROUGH ADDER MINIMIZATION TECHNIQUE AT DCT STRUCTURE
- IMAGE ANNOTATION BASED ON BAG OF VISUAL WORDS AND OPTIMIZED SEMI-SUPERVISED LEARNING METHOD
- VIDEO SEGMENTATION USING A NOVEL LBP DESCRIPTOR
- A VIDEO WATERMARKING ALGORITHM BASED ON TURBO CODES IN THE MULTI-TRANSFORM DOMAIN
Last modified: 2014-09-02 14:06:06