Facial feature extraction system and method based on FECNN

Abstract

The invention relates to a facial feature extraction system and a method based on FECNN. The system comprises a face pretreatment module, an FECNN module, a feature extraction module and a feature comparison module, wherein the face pretreatment module is used for carrying out face detection on a face image and carrying out cutting, positioning and aligning on the detected face image; the FECNN module builds an FECNN frame for feature extraction and carries out training until the FECNN is converged, and an FECNN parameter model is obtained; the feature extraction module sends face key points and the face image to the FECNN parameter model to extract face features; and the feature comparison module uses a cosine distance to calculate the face features, when the distance is larger than a set threshold t, the same person is judged, and when the distance is smaller than the set threshold t, different persons are judged. Compared with the prior art, the system and the method of the invention use few parameters, the network model can be quickly converged, and robust face features can be extracted.

Description

technical field [0001] The invention relates to the technical field of face recognition, in particular to a system and method for extracting face features based on FECNN. Background technique [0002] Face recognition is currently one of the hottest research topics in the fields of machine learning, pattern recognition and computer vision, and it has more and more applications in practical engineering. After decades of research, fruitful results have been obtained. In recent years, CNN has been used for image processing many times, and it has been proved that it is more practical and has higher accuracy than traditional methods in feature extraction of targets. In particular, CNN has made breakthrough progress in face recognition, showing unlimited prospects; but in the existing technology, the features extracted by the shallow CNN network cannot fully represent the face, and the deep CNN network Faced with too many iterations and too long convergence time. Contents of t...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to provide a kind of human face feature extraction system based on FECNN, the technical problem to be solved is: the feature that shallow CNN network extracts can't represent human face fully, and the CNN network of deep layer is faced with too many iterations , the convergence time is too long

Images