To search, Click below search items. |
| All Published Papers Search Service |
|
|
|
Title |
The Negative Selection Algorithm: a Supervised Learning Approach for Skin Detection and Classification. |
|
Author |
Esma Bendiab, Mohamed Kheireddine Kholladi |
| Citation |
Vol. 10 No. 11 pp. 86-92 |
|
Abstract |
Artificial immune systems (AIS) are relatively new class of meta-heuristics that mimics aspects of the human immune system to solve computational problems. They consist of three typical intelligent computational algorithms termed clonale selection algorithm, immune network theory and negative selection algorithm. The negative selection algorithm is a supervised learning algorithm based population. It has been successfully applied to change and anomaly detection. As AIS emerged, classification has been become an important application area of AIS. Classification systems that are based on AIS have attractive features inherited form biological immune system. On the other hand, skin color detection is used as a preliminary step in numerous computer vision applications like: face detection, hand gesture detection, person identification and others. Two main issues of the skin detection are: the selection of the best skin features space and the skin pixel classification algorithm. This paper describes initial framework of pixels skin color classification approach based on the negative selection algorithm from AIS. We present a skin classification, for detecting skin pixels and non skin pixels in color images, using both color and texture features space and the negative selection algorithm as a classifier. The proposed approach is able to detect skin regions from images taken from different imaging conditions. The results are promising and suggest a new approach for adapting human skin color using negative selection algorithm. |
|
Keywords |
Artificial Immune System, Negative Selection Algorithm, Classification, Supervised Learning, Skin Detection |
|
URL |
