人脸检测技术在“多媒体课程”中的教学应用  

Application of face detection technology in“multimedia course”

作　　者：张湃 孟庆莹 Zhang Pai;Meng Qingying(College of Intelligence and Information Engineering,Tangshan University,Tangshan Hebeid,063000)

机构地区：[1]唐山学院智能与信息工程学院,河北唐山063000

出　　处：《电子测试》2022年第7期138-140,共3页Electronic Test

基　　金：数字图像处理“金课”建设教学模式的创新与实践(JG21183)。项目负责人:张湃,单位:唐山学院,项目分类:教研项目-校级项目-唐山学院教学改革研究与实践项目。

摘　　要：设计了连续平均人脸检测综合性实验项目。由于连续平均量化算法在提取人脸特征方面,表现出对光照与传感器多变的不敏感特性,并且计算复杂度较低,因此用于作为人脸面部特征提取方法,并采用SNOW分类器进行分类。SNOW分类器是一种基于线性特征空间的的稀疏网络结构,可以针对人脸面对的各种特征、表情、以及光照不均和遮挡物等情况进行训练,并得到识别数据,其鲁棒性与识别准确率都较高。算法在MATLAB仿真平台上进行仿真。该实验设计涵盖了信号与系统、图像处理、线性代数、人工智能、数据分析等知识,具有综合性、适用性和可扩展性,将其实施在数学建模课程有助于提高学生的创新能力和实践技能。Digital image processing is a major professional course for electronics,communication and other majors.The course integrates the knowledge of the previous basic courses and expands its application.In order to deepen the educational reform of this course,strive to change the current teaching situation,so that students can not only have theoretical knowledge,but also benefit from cutting-edge and employment through the study of this course,a reform scheme based on“digital image processing”course is proposed,which plans to carry out reform from the aspects of bone stem teacher training,auxiliary teaching platform,bilingual teaching and so on to strengthen the interaction between teachers and students,Improve students’comprehensive ability,and design a comprehensive experimental design of image fusion.Cultivate students’practical ability.

关 键 词：人脸检测 连续平均量化 SNOW分类器 特征提取 

分 类 号：TP37[自动化与计算机技术—计算机系统结构]