Research On Automatic Quantitative Assessment Of Facial Paralysis

Facial Paralysis is a condition where damage to the facial nerve causes weakness of the muscles on one side of the face resulting in an inability to close the eye and dropping of the angle of the mouth. It brings both physical and mental Inconvenience to the patients.Grading of facial nerve plays an important role in treatment of facial paralysis. But most methods of assessing facial function are subjective and there is considerable variability in the results between different assessors. As image processing and pattern recognition grow fast, we can use them to develop an automated, objective and reliable facial grading system.The research of facial feature extraction is valuable and it becomes a focus in image processing and pattern recognition. This paper analyzes some methods of facial feature extraction and presents a facial grading system which includes image processing, detection of faces, feature extraction and comparability of features.This paper first introduces the background and significance of facial grading system, then talks about pre-processing and normalization of the image. It presents methods of using corner detection and feature extraction and compares them. Finally the paper introduces the application of SVM in facial grading system.The methods of this paper are supported by many experiments and are proved to be feasible.