How to use pca to realize face recognition in python

Face recognition method based on feature face

Feature face method is a face recognition method based on KL transform, which is an optimal orthogonal transform for image compression. A new set of orthogonal bases is obtained from the high-dimensional image space after KL transformation, and the important orthogonal bases are retained, thus a low-dimensional linear space can be generated. If it is assumed that the projections of human faces in these low-dimensional linear spaces are separable, then these projections can be used as feature vectors for recognition, which is the basic idea of feature face method. These methods need more training samples and are completely based on the statistical characteristics of image gray level. At present, there are some improved feature face methods.

For example, a gray-scale photo of human face has 40x40= 1600 pixels, and this person's face is represented by a matrix composed of gray-scale values of each pixel. Then everyone's face needs 1600 features. Taking a bunch of such samples as pca, only a few features that can represent a sample in statistical sense are extracted.

Face recognition can adopt the idea of neural network deep learning. There are many domestic cases of ColorReco.