一种高校后勤工作人员面部表情识别方法

A method of facial expression recognition for university logistics staff

为推进我国高校智慧校园建设,提高后勤人员的工作积极性,文中提出一种面向后勤工作人员的面部表情识别方法。该方法首先利用工作现场的摄像头、数据采集卡等,构建一种适合校园后勤工作场景的图像数据采集平台;然后对采集到的后勤工作人员面部表情数据进行直方图均衡化、Rudin-Osher-Fatemi去噪、图像矫正,并进行眼部表情图像数据的截取和预处理,以建立面向高校后勤工作人员的面部表情识别数据集。应用结果表明:所提方法的面部表情识别率达到92.17%,说明该方法能够反映高校后勤工作人员的工作情绪及工作积极性。

In order to promote the construction of smart campuses in Chinese universities and enhance the enthusiasm of logistics staff,a facial expression recognition method for logistics staff is proposed.In this method,on-site cameras,data acquisition cards,etc.are first utilized to construct an image data acquisition platform suitable for campus logistics work scenarios.The Histogram equalization,Rudin-Oser-Fatemi denoising and image correction are conducted on the collected facial expression data of logistics staff,and the eye expression image data is intercepted and preprocessed to establish a facial expression recognition dataset for university logistics staff.The application results show that the facial expression recognition rate of the proposed method can reach 92.17%,which indicates that this method can reflect the work emotions and enthusiasm of university logistics staff.