Face in-vivo detection system based on color and singular value features

Abstract

The invention provides a face in-vivo detection system based on color and singular value features, comprising an original data building module, a feature extraction module, a feature processing module, a training module, and a prediction module. The original data building module is used for dividing original data into preliminary samples and test samples. The feature extraction module is used for dividing the face images in the preliminary samples into blocks and extracting the color features and singular value features of the face image blocks in batch. The feature processing module is used for reducing the preliminary samples to get training samples. The training module is used for training a support vector machine by use of the training samples and the parameters of the support vector machine after optimization to get a face in-vivo detection model. The prediction module is used for predicting the features of the test samples using the face in-vivo detection model to get a classification result of a living or disguised person. Complicated extraction of features in the past is avoided, and the computational complexity is reduced greatly.

Description

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine

Login to view more

AI Technical Summary

Benefits of technology

This technology uses colors or specific values for facial recognition instead of complicated data structures that were previously used during previous methods like histograms analysis. It can detect faces without extracting any unnecessary detail from them while still being able to accurately identify their identity with high accuracy.

Problems solved by technology

This patented technology describes how we can use facial recognition techniques (FAC) to recognize faces that were previously unknown beforehand - this was done without any specific assumptions about what they really belong to like humans. However, there could be situations where these FAC algorithms mistakenly identified somes who weren't actually alive during an attack.

Images