Human face living body recognition method

Abstract

The human face living body recognition method comprises the steps of obtaining human face living body data and human face non-living body data; carrying out FFT on the face living body data and the face non-living body data respectively; obtaining a deep learning neural network model of human face living body recognition; training a deep learning neural network model of human face living body recognition by using the transformed human face living body data as a positive sample training set and using the transformed human face non-living body data as a negative sample training set to obtain a trained human face living body recognition neural network model; carrying out the face living body recognition by adopting trained face living body recognition neural network model, and judging whetherrecognized object is face living body or not. According to the face living body recognition method, high-precision face living body recognition is achieved with the small calculation amount and the small storage space, and the face living body recognition method has the advantages of being stable in operation, extremely low in power consumption, high in precision, good in user experience and thelike; the method is especially suitable for being applied to Internet of Things equipment, mobile terminal equipment and edge computing equipment.

Description

technical field [0001] The invention relates to the technical field of face recognition, in particular to a face recognition method. Background technique [0002] Face recognition is a biometric technology that identifies people based on their facial feature information. A series of related technologies such as using a camera to collect images / video streams containing faces, detecting and tracking faces in the collected images, and performing facial feature recognition on detected faces are also called portrait recognition, facial recognition, etc. [0003] Face features are easily imitated by photos, videos, 3D structures, etc. Therefore, in order to prevent fake face features from being used for face recognition system verification, the face recognition system needs to add a face recognition function, that is, to judge the submitted Whether the facial features come from a living real person. The face recognition function in the face recognition verification and authenti...

Images