Selective Ensemble Face Recognition Method Based on CSJOGA Algorithm

Abstract

The invention discloses a selective integration face recognition method based on the CSJOGA algorithm, which belongs to the technical field of face recognition, in particular a combined sparse projection-based improved hybrid orthogonal genetic algorithm (CSJOGA), bagging method, selective integration technology, multiple A new approach to face recognition with categorical SVM methods. This method combines minimum spanning tree clustering, joint sparse projection and orthogonal experimental design, and proposes an orthogonal intersection operator based on joint sparse projection. Then the crossover is designed by using the orthogonal experiment method, and a clustering local search strategy is introduced, and a hybrid orthogonal genetic algorithm based on joint sparse projection is proposed. This method can ensure that the correct rate of face recognition can be further improved to a higher level under the premise of greatly reducing the number of integrated classifiers, reducing storage and computing overhead, and effectively improving the speed of face recognition, ensuring that the classification accuracy in practical applications reaches Ideal requirements.

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