Lips detection using neural networks

Jamal Ahmad Dargham, and Ali Chekima, and Sigeru, Omatu (2008) Lips detection using neural networks. In: 13th International Symposium on Artificial Life and Robotics (AROB 13th'08), 31 January - 2 February 2008;, Oita,.

Full text not available from this repository.

Official URL: http://dx.doi.org/10.1007/s10015-007-0494-0

Abstract

Lips detection is used in many applications such as face detection and lips reading. In this paper a method for lips detection in colour images in the normalised RGB colour scheme is presented. In this method, MLP neural networks are used to perform lips detection on segmented skin regions. Several combinations of chrominance components of the normalized RGB color space were used as the input to the neural networks. Two methods were used for obtaining the normalized RGB components from the RGB colour scheme. These are called maximum and intensity normalization methods respectively. The method was tested on two Asian databases. The number of neurons in the hidden layer was determined by using a modified network growing algorithm. It was found out that the pixel intensity normalisation method gave lower lips detection error than the maximum intensity normalisation method regardless of the database used and for most of the combinations of chrominance components. In addition, the combination of the g and r/g chrominance components gave the lowest lips detection error when pixel intensity normalisation method is used for both databases. The effect of the scale and facial expression on the lips detection was also studied. It was found out that the lips detection error decreases as the scale factor increases. As for the facial expression, laughing facial expression gave the highest lips detection error followed by smiling and neutral expressions. ©ISAROB 2008.

Item Type:Conference Paper (UNSPECIFIED)
Uncontrolled Keywords:Colour image, Detection error, Face Detection, Facial Expressions, Hidden layers, Maximum intensities, MLP neural Colour image, Detection error, Face Detection, Facial Expressions, Hidden layers, Maximum intensities, MLP neural networks, Normalisation, Normalization methods, Normalized RGB color space, Pixel intensities; RGB colours, Scale Factor
Subjects:?? TR624-835 ??
Divisions:SCHOOL > School of Engineering and Information Technology
ID Code:1610
Deposited By:IR Admin
Deposited On:25 Mar 2011 09:18
Last Modified:30 Dec 2014 14:49

Repository Staff Only: item control page


Browse Repository
Collection
   Articles
   Book
   Speeches
   Thesis
   UMS News
Search
Quick Search

   Latest Repository

Link to other Malaysia University Institutional Repository

Malaysia University Institutional Repository