Facial paralysis grading comprehensive evaluation system combining facial textures and optical flow characteristics

Abstract

The invention discloses a facial paralysis grading comprehensive evaluation system combining facial textures and optical flow characteristics. The system comprises an image video acquisition and preprocessing module, a facial symmetry area division module, a texture difference-based evaluation module, an optical flow characteristic difference-based evaluation module and a facial paralysis gradingcomprehensive evaluation module. The method comprises the following steps: firstly, preprocessing a facial paralysis static image and video data; carrying out region division on the human face by utilizing a human face key point detection method; for the facial paralysis image and the video data, completing facial paralysis grading evaluation according to the texture feature difference and the optical flow feature difference of the corresponding areas on the left side and the right side of the face respectively; and finally, carrying out comprehensive evaluation of facial paralysis grading byutilizing an evaluation result based on images and video data. Experiments show that compared with a traditional method, the average accuracy rate of facial paralysis grading evaluation is increased by 18% or above, and the method has obvious advantages.

Description

technical field [0001] The invention relates to the technical fields of medical treatment and imaging, in particular to a facial paralysis grading comprehensive evaluation system combined with facial texture and optical flow features. Background technique [0002] Facial palsy is a disease characterized by motor dysfunction of facial expression muscles. It is a common disease, frequently occurring disease, and is not restricted by age. Facial paralysis is mainly caused by facial nerve damage, most of which are unilateral facial paralysis, mainly manifested in crooked mouth and eyes, unable to complete basic facial movements such as raising eyebrows, closing eyes, and puffing cheeks. When the face is moving, the difference between the left and right sides of the face is more obvious. At present, the diagnosis of facial paralysis mainly relies on the doctor's diagnostic method of "seeing, smelling, asking and feeling" to diagnose the patient's condition and evaluate the sever...

AI Technical Summary

Problems solved by technology

[0006] Aiming at the problems that the traditional methods only use facial texture and shape features to evaluate the grade of facial paralysis, and the characteristics of facial movement changes are not considered in deep learning, the present invention proposes a combination of facial texture A comprehensive evaluation system for the classification of facial paralysis based on differences in characteristics of optical flow and optical flow

Images