abstract |
The invention provides a convolutional neural network dynamic gesture authentication method, device, storage medium and device; wherein the method includes the following steps: setting a working mode to a registration mode or an authentication mode; inputting a user id; collecting a user's dynamic gesture video; The dynamic gesture video is preprocessed; it is input to the gesture feature extractor, and the feature vector containing the user identity information is extracted; in the registration mode, the input user id and the extracted feature vector are added to the registration feature library; in the authentication mode Next, calculate the cosine distance between the extracted feature vector and each feature vector corresponding to the input user id in the registered feature database; if the minimum cosine distance is less than the authentication threshold, the authentication is passed. The method can not only quickly extract the dynamic behavioral features of gestures, but also contain physiological features with high user distinguishability, which can improve the performance of gesture authentication, and has good gesture authentication accuracy and response speed. |