摘要
为了进一步提高人脸微笑检测率并解决微笑检测系统用于训练标签数据不足的问题,结合人脸图像的纹理和几何特点,应用了一种基于多分类模型加权投票法的微笑检测方法。在经预处理和直方图均衡化的面部图像上,利用局部二进制模式(LBP)和Gabor小波变换提取局部的、抽象的特征,同时以人脸特征点检测作为补充,构建了四种不同的分类模型(UPN、GPS、AdaBoost和LDA),分别对人脸图像进行分类检测,同时结合各模型之间互补和各自对微笑检测的优势,通过计算权值对各结果进行加权投票,得到面部图像的最终检测结果。实验结果显示出该方法的有效性,在公开的GEN-KI-4K人脸数据集上获得了95.8%的微笑检测率,比单个分类模型的平均检测率提高了10.3%,与该数据集的最新的微笑检测率相等。
In order to further improve the face smile detection rate and solve the problem that the smile detection system is not enough for training tag data,we apply a smile detection method based on multi-classification model weighted voting in combination with the texture and geometric characteristics of human face image.On the facial image with pre-processing and histogram equalization,local and abstract features are extracted by the local binary model(LBP)and Gabor wavelet transform.At the same time,facial feature point detection is used as a supplement to construct four different classification models(UPN,GPS,AdaBoost and LDA).Combined with the complementarity between the models and their respective advantages on smile detection,the final detection results of facial images are obtained by calculating the weights and voting the results weighted.The experiment shows the effectiveness of this method.The detection rate of 95.8% is obtained on the open GENKI-4K face dataset,which is 10.3% higher than the average detection rate of a single classification model and the same as the latest smile detection rate.
作者
冯泽安
王鹏
FENG Ze-an;WANG Peng(School of Electronic Information Engineering,Xi’an Technological University,Xi’an 710021,China)
出处
《计算机技术与发展》
2019年第2期81-86,共6页
Computer Technology and Development
基金
陕西省科技计划重大重点项目(2017ZDXM-GY-114)
关键词
微笑检测
人脸图像
纹理
几何
分类模型
加权投票法
smile detection
face image
texture
geometry
classification model
weighted voting
作者简介
冯泽安(1993-),男,硕士研究生,研究方向为图像处理;王鹏,副教授,研究方向为图像处理和模式识别。
