Motion detection facial paralysis degree evaluation system based on artificial intelligence

Abstract

The invention relates to the technical field of artificial intelligence, in particular to a motion detection facial paralysis degree evaluation system based on artificial intelligence. The system comprises: a key point detection module used for obtaining face key point information of a to-be-detected person; a classification module which is used for obtaining relevance scores according to the movement speed and the earliest movement time of the key points in the video of the health personnel under the specified movement, setting relevance score thresholds and dividing the active key points according to the relevance scores; a related point determining module which is used for acquiring a key point with relatively large motion amplitude in the active key points as a related point and constructing a set of the related points; a key frame screening module which is used for selecting a key frame in the video according to the total weight of the related points of each frame of image; and animage detection module which is used for acquiring target areas of the key frames under different specified actions and sending the target areas into the classification network to detect and output the severity level of facial paralysis. According to the invention, the related points and key frames of the facial paralysis patient in the facial movement process can be accurately determined, and the detection speed is improved.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to an artificial intelligence-based motion detection facial paralysis evaluation system. Background technique [0002] Facial paralysis is a common and frequently-occurring disease in our daily life. It is a nervous system disease, and it is difficult to treat. The common examination methods for facial paralysis are divided into static examination and exercise examination. Static examination means that the patient keeps still. Check whether the patient's facial features are normal, and exercise examination means that the doctor asks the patient to do some specific exercises to check whether the patient's facial functions are normal. [0003] The current intelligent facial paralysis severity detection mostly uses the combination of static detection and motion detection. The main problem it faces is that there are many key points on the face of the person to be detecte...

AI Technical Summary

Problems solved by technology

[0003] The current intelligent facial paralysis severity detection mostly uses the combination of static detection and motion detection. The main problem it faces is that there are many key points on the face of the person to be detected. During the motion detection process, it is necessary to analyze the motion state of all these key points. Moreover, due to the facial movement of the inspector, it is necessary to obtain multi-frame video information during the movement process, the amount of data is large, and the calculation process is relatively complicated, resulting in a slow detection process

Images