Face recognition method based on image feature analysis

Abstract

The invention discloses a face recognition method based on image feature analysis. The method comprises the steps of 1, processing the image of a face to be recognized, extracting scale-invariant feature transform (SIFT) feature point vector, taking the SIFT feature point vector as basis of recognition and judgment, and carrying out rotating optimization on the extracted SIFT feature point vector; 2, reducing the dimensionality of the extracted SIFT feature point vector; 3, selecting the SIFT feature points of the image, and matching the feature points. After the method based on feature analysis and matching is adopted, the dimensionality of the feature vector can be reduced, the image matching effect is not influenced when the number of the feature points is reduced, the speed and the accuracy of face recognition are improved, the robustness of recognition can be improved, and the real-time online running speed and the matching accuracy of the whole method are improved.

Description

technical field [0001] The invention relates to a face recognition method, in particular to a face recognition method based on image feature analysis. Background technique [0002] Among the methods for appearance model construction and recognition, one of the important methods is mainly divided into two methods: global appearance and local appearance recognition. The core embodiment is the extraction of well-known global features and local features. In face recognition, handwritten Chinese character recognition and other recognition fields, feature extraction is one of the most important links. In the 1980s, two scientists, Movarac and Hannah, proposed an algorithm about corner points, and then according to this algorithm, Stephens and Harris transformed this initial idea into the characteristic problem of the two eigenvalues ​​of the structure tensor 8 years later. , which can also be called a second-order matrix problem. Afterwards, scientists such as Rohr, Kanade, Toma...

AI Technical Summary

Problems solved by technology

However, global features generally can only perform rough matching, while local features can provide finer confirmation; recognition methods based on algebraic features represent human faces as feature vectors, forming a vector subspace called eigenfaces, in the space The face is projected, and then each corresponding feature coordinate coefficient is obtained. This set of coordinate coefficients reflects the position of the face, so that the position where the face exists can be found very stably to achieve the purpose of recognition, but because the feature face is easily affected by external factors Influences, such as: lighting, shooting angle, etc., which lead to a decline in the recognition effect; the recognition method based on geometric features expresses the face as a geometric feature vector, and combines the classifier design idea of ​​​​hierarchical clustering for recognition

Geometric-based feature vectors are produced by a certain relationship between the shape of the face and the geometry, and its components usually include the Euclidean distance, angle, curvature, etc. of two points on the face. The contours of organs and the search for feature points are relatively good, but they tend to be generalized, replacing the picture features of the entire face with a few characteristic phenomena, ignoring the details of the face, resulting in low recognition accuracy

Images