[1]王 珏,潘沛生.基于超分辨率重建的低分辨率表情识别的研究[J].计算机技术与发展,2021,31(07):47-51.[doi:10. 3969 / j. issn. 1673-629X. 2021. 07. 008]
　WANG Jue,PAN Pei-sheng.Research on Low-resolution Facial Expression Recognition Based on Super-resolution Reconstruction[J].,2021,31(07):47-51.[doi:10. 3969 / j. issn. 1673-629X. 2021. 07. 008]
点击复制

基于超分辨率重建的低分辨率表情识别的研究()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
31
期数:
2021年07期
页码:
47-51
栏目:
图形与图像
出版日期:
2021-07-10

文章信息/Info

Title:
Research on Low-resolution Facial Expression Recognition Based on Super-resolution Reconstruction
文章编号:
1673-629X(2021)07-0047-05
作者:
王 珏; 潘沛生
南京邮电大学 通信与信息工程学院,江苏 南京 210003
Author(s):
WANG Jue; PAN Pei-sheng
School of Telecommunications and Information Engineering,Nanjing University of Posts and Telecommunications, Nanjing 210003,China
关键词:
人脸表情识别; 超分辨率重建; 混合损失函数; 低分辨率图像; 小尺度卷积核
Keywords:
facial expression recognition;  super-resolution reconstruction;  joint loss function;  low resolution image;  small- scale convolution kernel
分类号:
TP391. 41
DOI:
10. 3969 / j. issn. 1673-629X. 2021. 07. 008
摘要:
人脸表情识别一直是人机交互和计算机视觉研究的热点问题, 但受限于成像设备及环境的影响,获得的面部图像往往是低分辨率的。 针对这个问题,提出一种基于深度学习的超分辨率重建的人脸表情识别系统,用于提高低分辨率面部图像表情识别的准确率。 该系统是由两个深度神经网络组成,首个神经网络基于新的混合损失函数,利用简化的残差网络结构叠加构成残差块来学习面部图像的特征,并进行上采样操作, 重建包含更多细节的高分辨率人脸表情图像; 第二个神经网络通过使用小尺度卷积核提取重建后的高分辨率图像的人脸表情特征,之后使用 softmax 分类器,? 实现人脸表情分类。 在公共数据集 Cohn-Kanade Dataset (CK+)上对该系统进行验证和测试的结果表明,该系统有效地提高了不同尺寸的低分辨率图像表情识别的准确率。
Abstract:
Facial expression recognition has always been a hot issue in human-computer interaction and computer vision research,but the facial images obtained are often low-resolution due to the impact of imaging equipment and the environment. To solve this problem,a facial expression recognition system based on deep learning super-resolution reconstruction is proposed to improve the accuracy of facial expression recognition in low-resolution facial images. The system is composed of two deep neural networks. The first neural network uses a simplified residual network structure to form a residual block to learn the features of the facial image,and performs an up-sampling operation based on a new hybrid loss function to reconstruct more detailed high-resolution facial expression images. The second one uses a small-scale convolution kernel to extract the facial expression features of the reconstructed high-resolution image,and then uses the softmax classifier to realize facial expression classification. The proposed system is verified and tested on the public data set Coh-nKanade Data set (CK+),which shows that this system improves the accuracy of the recognition of low-resolution expressions of different sizes effectively.

相似文献/References:

[1]王春霞 苏红旗 范郭亮.图像超分辨率重建技术综述[J].计算机技术与发展,2011,(05):124.
　WANG Chun-xia,SU Hong-qi,FAN Guo-liang.Overview on Super Resolution Image Reconstruction[J].,2011,(07):124.
[2]潘峥嵘,贺秀伟.人脸表情识别在智能机器人中的应用研究[J].计算机技术与发展,2018,28(02):173.[doi:10．3969/j．issn．1673－629X．2018．02．037]
　PAN Zheng-rong,HE Xiu-wei.Research on Application of Facial Expression Ｒecognition in Intelligent Ｒobot[J].,2018,28(07):173.[doi:10．3969/j．issn．1673－629X．2018．02．037]
[3]崔凤焦.表情识别算法研究进展与性能比较[J].计算机技术与发展,2018,28(02):145.[doi:10．3969/j．issn．1673－629X．2018．02．031]
　CUI Feng-jiao.Ｒesearch and Performance Comparison of Facial Expression Ｒecognition Algorithm[J].,2018,28(07):145.[doi:10．3969/j．issn．1673－629X．2018．02．031]
[4]沈怡灵,赵明哲,李强懿,等.基于稀疏表示的二值图像超分辨率重建算法[J].计算机技术与发展,2017,27(12):43.[doi:10.3969/ j. issn.1673-629X.2017.12.010]
　SHEN Yi-ling,ZHAO Ming-zhe,LI Qiang-yi,et al.A Super-resolution Reconstruction Algorithm for Binary Image Based on Sparse Representation[J].,2017,27(07):43.[doi:10.3969/ j. issn.1673-629X.2017.12.010]
[5]史国川,龚连友.基于系数复用和字典训练的图像超分辨率算法[J].计算机技术与发展,2018,28(03):114.[doi:10.3969/ j. issn.1673-629X.2018.03.024]
　SHI Guo-chuan,GONG Lian-you.An Image Super-resolution Reconstruction Algorithm Based on Coefficient Reuse and Dictionary Training[J].,2018,28(07):114.[doi:10.3969/ j. issn.1673-629X.2018.03.024]
[6]俞文静,张明军,王影.面向视频超分辨率重建的混合粒子群优化算法[J].计算机技术与发展,2018,28(11):75.[doi:10.3969/ j. issn.1673-629X.2018.11.017]
　YU Wen-jing,ZHANG Ming-jun,WANG Ying.A Hybrid Particle Swarm Optimization Algorithm for Image/ Video Super-resolution Reconstructio[J].,2018,28(07):75.[doi:10.3969/ j. issn.1673-629X.2018.11.017]
[7]张雪梅,公维宾,邬建志,等.基于纹理特征融合的人脸表情识别[J].计算机技术与发展,2020,30(03):57.[doi:10. 3969 / j. issn. 1673-629X. 2020. 03. 011]
　ZHANG Xue-mei,GONG Wei-bin,WU Jian-zhi,et al.Facial Expression Recognition Based on Texture Feature Fusion[J].,2020,30(07):57.[doi:10. 3969 / j. issn. 1673-629X. 2020. 03. 011]
[8]周传华,吴幸运,李 鸣.基于 WGAN 单帧人脸图像超分辨率算法[J].计算机技术与发展,2020,30(09):29.[doi:10. 3969 / j. issn. 1673-629X. 2020. 09. 006]
　ZHOU Chuan-hua,WU Xing-yun,LI Ming.Single Frame Face Images Super-resolution Algorithm Based on WGAN[J].,2020,30(07):29.[doi:10. 3969 / j. issn. 1673-629X. 2020. 09. 006]
[9]付倩倩,李 昂.一种改进的卷积神经网络的表情识别算法[J].计算机技术与发展,2020,30(11):80.[doi:10. 3969 / j. issn. 1673-629X. 2020. 11. 015]
　FU Qian-qian,LI Ang.An Improved Facial Expression Recognition Technology Based on Convolutional Neural Network[J].,2020,30(07):80.[doi:10. 3969 / j. issn. 1673-629X. 2020. 11. 015]
[10]植炜基,刘春雨,郑婉君,等.基于生成对抗网络的人脸表情识别技术综述[J].计算机技术与发展,2021,31(增刊):1.[doi:10. 3969 / j. issn. 1673-629X. 2021. S. 001]
　ZHI Wei-ji,LIU Chun-yu,ZHENG Wan-jun,et al.Survey of Facial Expression Recognition Technology Based onGenerative Adversarial Network[J].,2021,31(07):1.[doi:10. 3969 / j. issn. 1673-629X. 2021. S. 001]

更新日期/Last Update: 2021-07-10