An identity verification method of biological characteristics based on a hand-painted track

The invention discloses an identity verification method of biological characteristics based on a hand-painted track. The method comprises the following steps: collecting a track training image; Preprocessing the collected track training image to obtain a training sample; Constructing a convolutional neural network model; Training the convolutional neural network model based on the training sampleto obtain a track training model; Acquiring a track password set by a user, and training the track training model based on the track password to obtain a track password verification model of the user;And obtaining a to-be-verified track password input by the user, and verifying the to-be-verified track password based on the track password verification model of the user. According to the identityverification method provided by the invention, a user can carry out unlocking by using a password to realize identity verification and similar scenes; The method has the advantages that only simple tracks need to be drawn, trac