Human face authentication method and device

Abstract

The invention discloses a human face authentication method, which comprises the following steps of: extracting human face characteristics from a plurality of frame images; judging whether the human face characteristics in the plurality of frame images accord with a preset human face model or not, and when the number of images according with the preset human face model exceeds a first threshold value, considering that user authentication corresponding to the human face characteristics in the plurality of frame images is successful; when the number of images according with the preset human face model exceeds a second threshold value, extracting all human face characteristics in the plurality of frame images as sample characteristics; and performing increment training on the preset human face model by utilizing the sample characteristics, wherein the second threshold value is not less than the first threshold value. In the method, more approximate sample images are used for training the human face model in the continual using process of users, so that the authentication capacity of the human face model is enhanced, and the self-maintenance is realized.

Description

【Technical field】 [0001] The invention relates to the technical field of image processing, in particular to a face authentication method and device capable of incremental learning. 【Background technique】 [0002] In the field of image processing and computer vision technology, face authentication technology is a relatively important and mature technology. The existing face authentication technology usually adopts the method based on the training method to obtain the face model. Specifically, the face area is cut from the training image and the features are extracted as samples, and a certain training method is used to train the model to be able to distinguish the samples. classifier, and then use the classifier for face authentication. [0003] However, the face model in this method is only obtained by training an initial batch of sample images. Factors such as changes in time, changes in ambient light, changes in clothing, and changes in appearance will cause the success ...

AI Technical Summary

Problems solved by technology

But as time changes, in winter, Zhang San changed a new hairstyle and wore more clothes, and the authentication rate of the attendance machine will slowly decline.

At this time, the traditional method is to re-acquire training images and train models for Zhang San by the attendance machine, because this method requires additional manual operations, increases the maintenance cost of the system, and the user experience is also very bad

Images