Title :
A new quantitative assessment method of facial paralysis based on motion estimation
Author :
Cui, Yaoyao ; Wan, Mingxi ; Li, Junbo
Author_Institution :
Dept. of Biomed. Eng., Xi´´an Jiaotong Univ., China
fDate :
29 Oct-1 Nov 1998
Abstract :
A novel method is developed to assess facial expression motion and paralysis quantitatively using an optical flow technique based on genetic algorithm (GAOF). This method is more reliable than other assessment methods when only small subpixel motions occur. Experimental results demonstrate that this method is very useful to diagnose the site of facial paralysis and assess progression or recovery profiles of patients combined with other diagnosis methods
Keywords :
feature extraction; genetic algorithms; image sequences; medical image processing; motion estimation; video signal processing; dynamic image sequences; facial expression motion; facial paralysis; feature preferences; genetic algorithm; mark placing rules; motion estimation; optical flow technique; paralysis site; progression profiles; quantitative assessment method; recovery profiles; region-specific motion; small subpixel motions; Biomedical engineering; Biomedical optical imaging; Cameras; Eyes; Face detection; Genetic algorithms; Image motion analysis; Image sequences; Motion estimation; Optical computing;
Conference_Titel :
Engineering in Medicine and Biology Society, 1998. Proceedings of the 20th Annual International Conference of the IEEE
Conference_Location :
Hong Kong
Print_ISBN :
0-7803-5164-9
DOI :
10.1109/IEMBS.1998.747147
