Wyniki wyszukiwania - BazTech

Ograniczanie wyników

2 Machine Graphics and Vision

2 2009

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Sortuj według:

Ogranicz wyniki do:

Generalized Gaussian Density for Skin Detection in DCT Domain

Ghouzali S., Hemami S., Rziza M., Aboutajdine D.

Machine Graphics and Vision

2009

Vol. 18, No. 2

201-213

In this paper, we propose a highly efficient algorithm to model the human skin color. The algorithm involves generating a discrete Cosine transform (DCT) at each pixel location, using the surrounding points. The DCT coefficients incorporate the pixel color and texture information to distinguish between skin and non-skin. A generalized Gaussian distribution (GGD) is used in this framework to model the DCT coefficients at low frequencies. Next, the model parameters are estimated using the maximum-likelihood (ML) criterion applied to a set of training skin samples. Finally, each pixel is classified as skin if its likelihood ratio exceeds some threshold. The experimental results show that our model avoids excessive false detection while still retaining a high degree of correct detection.

An Improved Method for Face Recognition Based on SVM in Frequency Domain

Amine A., Ghouzali S., Rziza M., Aboutajdine D.

Machine Graphics and Vision

2009

Vol. 18, No. 2

187-199

We examine the problem of discriminating between objects of more than two classes using "minimum information". Discrete Cosine Transforms (DCT) represents a computationally simple and efficient method that preserves the structure of the data without introducing significant distortion. In this paper, an efficient face recognition method combined DCT and Support Vector Machine (SVM) is proposed. The underlying algorithm is derived by applying DCT to several regions of a face image. Only a small subset of the DCT coefficients is retained by truncating high frequency DCT components in each block. Selected DCT features are then subjected to SVM for class separability enhancement before being used for face recognition. This leads to a new, low-dimensional representation of images which allows for a fast and simple classification. In this context, we have performed a large number of experiments using two popular face databases: ORL and Yale, and comparisons using PCA, LDA, ICA, MLP, etc. Experimental results show that the proposed method performs better than traditional approaches in terms of both efficiency and accuracy.