摘要
针对传统AdaBoost算法在单特征分类器训练时耗费时间长、弱分类器质量低的问题,本文提出一种基于双特征的改进型AdaBoost分类检测算法。首先,通过粒子群寻优算法(PSO)搜寻最优的两个特征,以及两特征对应的阈值,形成双特征型弱分类器。接着将弱分类器组合成强分类器,最后在MATLAB软件中利用MIT人脸数据库进行仿真实验,结果表明本文基于双特征的分类器性能优于单特征分类器。
Aiming at the problems that the traditional AdaBoost algorithm takes a long time to train a single feature classifier and the quality of the weak classifier is low,this paper proposes an improved AdaBoost classification detection algorithm based on dual features.First,a particle swarm optimization algorithm(PSO)is used to search for the two optimal features and the thresholds corresponding to the two features to form a dual-feature weak classifier.Then,the weak classifiers are combined into a strong classifier.Finally,the MIT face database is used for simulation experiments in MATLAB software.The results show that the dual-feature-based classifier performs better than the single-feature classifier.
作者
张均
叶庆卫
ZHANG Jun;YE Qing-wei(College of Information Science and Engineering,Ningbo University,Ningbo 315211,China)
出处
《无线通信技术》
2020年第2期23-27,共5页
Wireless Communication Technology
基金
国家自然科学基金资助项目(No.51675286,No.61071198)。
