摘要
研究主要探讨基于分布式并行多遗传算法的学校公寓学生考勤人脸识别系统的优化问题。研究方法采用改进的分布式并行遗传算法,结合自适应的交叉和变异概率,通过实验对多个特征对识别精度的影响进行评估。研究结果表明,经过优化算法处理后,识别系统的错误率在不同特征下有所降低,脸型特征错误率为12.58%,眼睛大小错误率为15.99%,下巴轮廓错误率为9.88%,皮肤纹理错误率为14.50%。通过1 000个样本的训练,研究验证了提出的算法,其准确率最高达94.5%。研究的意义在于提高了学校公寓门禁系统的考勤管理智能化水平,为未来的系统实施与优化提供了科学依据。
The study mainly discusses the optimization problem of the school dormitory student attendance face recognition system based on distributed parallel multi-genetic algorithm.The research method adopts an improved distributed parallel genetic algorithm,combined with adaptive crossover and mutation probability,and evaluates the influence of multiple features on recognition accuracy through experiments.The research results show that after the optimization algorithm is processed,the error rate of the recognition system is reduced under different features.For example,the error rate of face shape features is 12.58%,the error rate of eye size is 15.99%,the error rate of chin contour is 9.88%,and the error rate of skin texture is 14.50%.Through the training of 1000 samples,the study verifies the proposed algorithm,and its accuracy rate is as high as 94.5%.The significance of the research is to improve the level of intelligent attendance management of the school dormitory access control system,and provide a scientific basis for future system implementation and optimization.
作者
王海燕
WANG Haiyan(Shaanxi Railway Institute,Weinan Shaanxi 714000,China)
出处
《自动化与仪器仪表》
2025年第7期184-189,共6页
Automation & Instrumentation
基金
陕西铁路工程职业技术学院辅导员精品项目《“三级四能”高职院校学生干部培养体系探索与实践》(2024fd-03)。
关键词
人脸识别
学生考勤管理
分布式并行多遗传算法
门禁系统
智能化管理
face recognition
Student attendance management
distributed parallel multi-genetic algorithm
access control system
intelligent management
