基于DCGAN的课堂表情图像生成方法
An Image Generation Method of Classroom Expression Images
徐新爱 1李钢1
作者信息
- 1. 南昌师范学院数学与信息科学学院,江西 南昌 330032
- 折叠
摘要
为了构建课堂表情图像数据库,弥补特定条件下课堂表情多样性的不足,提出一种利用深度卷积生成对抗网络(Deep Convolutional Generative Adversarial Networks,DCGAN)生成课堂表情图像的方法.首先,利用线下教学监控视频和线上课堂视频自主采集课堂表情图像,得到较均衡且样本特征丰富的小型图像集;其次,对原始图像进行去雾、增强、镜像等图像预处理操作,构建课堂表情数据训练集;再次,通过对基于DCGAN模型的课堂表情图像生成网络的构建和初步参数设置,并不断优化网络超参数,以生成课堂表情图像数据集;最后,利用人脸检测算法和IS(Inception Score)评价指标对生成课堂表情图像进行检测和评价,并验证生成图像在检测网络中的可行性和有效性.实验结果表明:本文基于DCGAN的方法能够生成较逼真的课堂表情图像,能够有效地增广课堂表情数据集,增强课堂表情图像的多样性.
Abstract
In order to build a database of classroom expression images and make up for the lack of classroom expression diversity under specific conditions,a method for generating classroom expression images based on deep convolutional generative adver-sarial networks(DCGAN)is proposed.Firstly,by using the offline teaching surveillance videos and the online classroom videos to independently collect classroom expression images,and a balanced and small image set with abundant sample features is ob-tained.Secondly,the training image set of classroom expression is constructed by image denoising,image enhancing and image mirroring.Thirdly,through the construction and preliminary parameter setting of the classroom expression image generation net-work based on DCGAN model,and constantly optimizing the network hyperparameters,the classroom expression image dataset is generated.Finally,the face detection algorithm and the IS(Inception Score)evaluation index are used to detect and evaluate the generated classroom expression images,and verify the feasibility and effectiveness of the generated images in the detection network.The experimental results show that the method based on DCGAN can generate more realistic classroom expression im-ages,effectively improve the classroom facial expression dataset,and enhance the diversity of classroom expression images.
关键词
深度学习/深度卷积生成对抗网络/图像生成/课堂表情Key words
deep learning/DCGAN(deep convolutional generation adversarial networks)/image generation/classroom expression引用本文复制引用
基金项目
国家自然科学基金专项项目(62341207)
江西省高校人文社会科学研究项目(JC20121)
江西省高校人文社会科学一般项目(JY21103)
出版年
2024
国家科技期刊平台
NSTL
万方数据