基于深度学习与随机森林的人脸图像年龄估计方法  

Age Estimation Method of Face Image Based on Deep Learning and Random Fores

作　　者：马迎辉 郝莉萍 李晓丹 MA Yinghui;HAO Liping;LI Xiaodan(College of Computer Science,Zhengzhou University of Economics and Business,Zhengzhou Henan 451191,China)

机构地区：[1]郑州经贸学院计算机学院,河南郑州451191

出　　处：《信息与电脑》2024年第6期35-37,共3页Information & Computer

基　　金：郑州经贸学院青年科研基金项目“基于深度森林的年龄估计”(项目编号:QK2216);郑州经贸学院青年科研基金项目“基于改进机器学习的无线通信网络欺骗流量攻击识别”(项目编号:QK2308);郑州经贸学院青年科研基金项目“基于强化学习的共享单车调度策略研究”(项目编号:QK2302)。

摘　　要：传统的人脸图像年龄估计方法只利用人脸图像中的部分信息,导致年龄估计的准确性较低。对此,提出基于深度学习与随机森林的人脸图像年龄估计方法:预处理人脸图像,提高图像辨识度并排除非人脸窗口;基于深度学习提取人脸图像特征,获取人脸图像信息;建立年龄标签分布编码,将提取的人脸图像特征转换为一种编码表示;利用随机森林模型估计年龄。实验结果表明,所提方法累积误差较小,精度较高。The traditional age estimation method of face image only uses part of the information in face image,which leads to the low accuracy of age estimation.Therefore,an age estimation method of face image based on deep learning and random forest is proposed:Preprocess the face image,improve the image recognition and side by side the non-face window;Facial image features are extracted based on deep learning,and facial image information is obtained;The age label distribution coding is established,and the extracted facial image features are transformed into a coded representation;The random forest model is used to estimate the age.The experimental results show that the cumulative error of the proposed method is small and the accuracy is high.

关 键 词：深度学习 随机森林 人脸图像 年龄估计 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]