An accurate and robust lip region detection algorithm based on skin and lip color segmentation is presented in this paper. Here skin and lip color analysis performs in chromatic and YCbCr color space respectively. The proposed algorithm defines geometrical models for skin and lip color distribution in order to detect skin and lip pixels in color image. The proposed algorithm performs lip detection in two stages. First skin pixels are detected in a given color image and face candidates are extracted. Then lip pixels are investigated within face candidates. We propose a segmentation method based on horizontal and vertical accumulation of the image pixel values. Experimental results show that the proposed method detects lips effectively with high accuracy.
Skin detection is the process of finding skin pixels in a given color image. In this paper, we propose a novel fuzzy rule based system for robust and fast skin detection in color images. The proposed fuzzy system models the skin color distribution in HSV color space. Using the fuzzy model, skin likelihood of the pixel is calculated and if it exceeds a threshold then this pixels is considered as skin pixel. Also we proposed a multi layer perceptron neural network for skin classifier and compare its performance with the proposed fuzzy rule based method.Experimental results show effectiveness of the proposed fuzzy rule based method in comparison with the neural network based and other previous approaches.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.