Title :
Lip contour extraction of RGB-based improved region growing algorithm
Author :
Chengjia Yang ; Xiongfei Li ; Xiaoli Zhang
Author_Institution :
Coll. of Comput. Sci. & Technol., Jilin Univ., Changchun, China
Abstract :
When one is speaking, lips change regularly. Therefore, Lip contour extraction is very meaningful. In this paper, for lip color images, this paper extracts lip contour by means of the RGB-based improved region growing algorithm. Through RGB three components, the relationship between color from and adjacent pixels is calculated. According to certain rules, seed pixels are automatically selected, and regional growth and de-noising and smoothing with median filtering are implemented. And finally, watershed algorithm is applied to get lip contour. Experimental results show that this method is simple to achieve the desired effect and good performance.
Keywords :
edge detection; image colour analysis; image denoising; median filters; RGB-based improved region growing algorithm; image denoising; image smoothing; lip color images; lip contour extraction; median filtering; Color; Computational modeling; Educational institutions; Feature extraction; Image color analysis; Lips; Vectors; color space; contour extraction; lips; region growing;
Conference_Titel :
Software Engineering and Service Science (ICSESS), 2014 5th IEEE International Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4799-3278-8
DOI :
10.1109/ICSESS.2014.6933639