Selective integration face identification method based on CSJOGA

Abstract

The invention discloses a selective integration face identification method based on a CSJOGA, belongs to the technical field of face identification, and particularly relates to a face identification new method based on a Combined Sparse Projection Improved Hybrid Orthogonal Genetic Algorithm (CSJOGA), a bagging method, a selective integration technology and a multi-classification SVM. The method combines minimum spanning tree clustering, combined sparse projection and orthogonal experimental design, and proposes an orthogonal crossover operator based on combined sparse projection. Then an orthogonal experimental method is utilized to design crossover, a clustering local search strategy is introduced, and a hybrid orthogonal genetic algorithm based on combined sparse projection is proposed. The method can ensure that the accuracy of face identification can be further improved to a higher level on the premise of substantially reducing the number of integrated classifiers, reducing storage and calculation overheads and effectively improving face identification speed, and ensure that classification precision in practical application reaches an ideal requirement.

Description

technical field [0001] The invention belongs to the technical field of face recognition, in particular to a new face recognition method based on joint sparse projection improved hybrid orthogonal genetic algorithm (CSJOGA), bagging method, selective integration technology and multi-classification SVM method. Background technique [0002] In recent decades, biometric recognition has received extensive attention in various fields, and has become one of the research hotspots in the field of pattern recognition and machine vision. Among them, face recognition is an important part of biometric recognition. [0003] In face recognition, face feature extraction is one of the key steps. Face feature extraction refers to extracting identification information that is helpful for classification from high-dimensional face data, and removing useless redundant information. [0004] In most cases, face features are divided into global features and local features. Among them, the global ...

AI Technical Summary

Problems solved by technology

[0008] Although genetic algorithms have made great progress in theoretical research, there are still a series of problems such as slow convergence speed, low search precision and "premature convergence" in practice and application.

In addition, in practical applications, there are often many factors that affect the optimization problem. If the genetic algorithm is used to solve it, a long code length is required, which will slow down the convergence of the algorithm.

Images