True and false facial paralysis identification system based on depth difference feature

Abstract

The invention discloses a true and false facial paralysis identification system based on depth difference features. The system comprises a training image acquisition module, which is used for acquiring training images and establishing a training image set; an identification network establishing module, configured to establish an identification network, and train the identification network by usingthe training image set to obtain an identification model; The identification network extracts the difference information of the deep features of the input image through the dual-branch convolution neural network, and then extracts the depth features by using the difference information of the single-branch convolution neural network to identify the true or false facial paralysis according to the difference of the depth features. The recognition module is used for acquiring an image to be recognized, and performing recognition to obtain a recognition result. The invention has good recognition effect, can effectively complete the recognition of true and false facial paralysis, and has important practical application value in clinical diagnosis.

Description

technical field [0001] The invention relates to the technical fields of medical treatment and image recognition, in particular to a real and fake facial paralysis recognition system based on depth difference features. Background technique [0002] Facial paralysis is a common disease with a wide range of incidence and is not limited by age. It will not only affect the life of the patient to a certain extent, but also cause a certain blow to his heart, seriously affecting the physical and mental health of the patient. With the increasing incidence of facial paralysis, more and more scholars have begun to pay attention to the identification of facial paralysis. [0003] In order to realize the automatic identification of facial paralysis, many scholars at home and abroad have conducted research on this aspect. They focus on static facial asymmetry and dynamic facial changes, track the movement differences of key points, and use deep learning methods to locate key points. Meth...

AI Technical Summary

Problems solved by technology

Because the researchers ignored the existence of false facial paralysis, the existing facial paralysis recognition methods have misjudgments, and there is a certain deviation in the accuracy of the recognition results, which reduces the accuracy of facial paralysis recognition to a certain extent.

Images