计算机科学 ›› 2022, Vol. 49 ›› Issue (12): 312-318.doi: 10.11896/jsjkx.211200036
孙长迪, 潘志松, 张艳艳
SUN Chang-di, PAN Zhi-song, ZHANG Yan-yan
摘要： 近年来,随着无人驾驶、无人机智能化、移动互联网的发展,低功耗、低算力的移动和嵌入式平台对轻量化的神经网络需求日益迫切。文中基于可变形卷积和深度可分离卷积思想的启发,提出了一种低开销可变形卷积,其兼具了可变形卷积的高效特征提取能力和深度可分离卷积的低运算量的优点。此外,在应用低开销可变形卷积的基础上,结合模型结构压缩的方法,设计了4种MobileNet网络再轻量化的方法。在Caltech256,CIFAR100和CIFAR10数据集上进行了实验,结果表明,低开销可变形卷积在运算量增加不明显的情况下,可以有效提高轻量级网络的分类准确度。并且,结合所提出的4种MobileNet再轻量化方法,可以将MobileNet网络的准确度提高0.4%~1%,与此同时网络运算量可以下降5%~15%,即显著提高了轻量化网络的各项性能,更加符合低功耗、低算力的现实需求,对于移动和嵌入式平台领域的智能化推进有着很重要的现实意义。
中图分类号:
| [1]SIMONYAN K,ZISSERMAN A.Very Deep Convolutional Networks for Large-Scale Image Recognition[J].arXiv:1409.1556,2014. [2]HE K,ZHANG X,REN S,et al.Deep Residual Learning forImage Recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).IEEE,2016. [3]PENGCHENG B I,LUO J,CHEN W,et al.Research on Lightweight Convolutional Neural Network Technology[J/OL].http://en.cnki.com.cn/Article_en/CJFDTotal-JSGG201916005.htm. [4]CHENG Y,WANG D,ZHOU P,et al.A Survey of Model Compression and Acceleration for Deep Neural Networks[J].arXiv:1710.09282,2017. [5]IANDOLA F N,HAN S,MOSKEWICZ M W,et al.Squeeze-Net:AlexNet-level accuracy with 50x fewer parameters and <0.5 MB model size[J].arXiv:1602.07360,2016. [6]HOWARD A G,ZHU M,CHEN B,et al.MobileNets:Efficient Convolutional Neural Networks for Mobile Vision Applications[J].arXiv:1704.04861,2017. [7]ZHANG X,ZHOU X,LIN M,et al.ShuffleNet:An Extremely Efficient Convolutional Neural Network for Mobile Devices[J].arXiv:1707.01083,2017. [8]DENG L,LI G,HAN S,et al.Model Compression and Hardware Acceleration for Neural Networks:A Comprehensive Survey[J].Proceedings of the IEEE,2020,108(4):485-532. [9]DAI J,QI H,XIONG Y,et al.Deformable Convolutional Net-works[C]//Proceedings of the IEEE International Conference on Computer Vision.Venice,2017:764-773.. [10]CHEN F,WU F,XU J,et al.Adaptive deformable convolutional network[J/OL].https://www.researchgate.net/publication/346289132_Adaptive_deformable_convolutional_network. [11]ZUO Z,ZHANG W,ZHANG D,et al.A remote sensing image semantic segmentation method by combining deformable convolution with conditional random fields[J].Acta Geodaetica et Cartographica Sinica,2020,3(3):39-49. [12]ZHANG Y F.Lightweight of deep learning and its application in image recognition[D].Jinhua:Zhejiang Teachers’ University,2019. [13]HE Y,ZHANG X,SUN J.Channel Pruning for AcceleratingVery Deep Neural Networks[C]//2017 IEEE International Conference onComputer Vision(ICCV).2017:1398-1406. [14]GUO J,LI Y,LIN W,et al.Network Decoupling:From Regular to Depthwise Separable Convolutions[J].arXiv:1808.05517,2018. [15]SUN W,ZHOU X,ZHANG X,et al.A Lightweight NeuralNetwork Combining Dilated Convolution and Depthwise Separable Convolution[C/OL]// 2020.https://link.springer.com/chapter/10.1007/978-3-030-48513-9_17. [16]GRIFFIN G,HOLUB A,PERONA P.Caltech-256 Object Category Dataset[R].Pasadena(Los Angeles,USA):Caltech Technical Report,2007. |
| [1] | 武红鑫, 韩萌, 陈志强, 张喜龙, 李慕航. 监督和半监督学习下的多标签分类综述 Survey of Multi-label Classification Based on Supervised and Semi-supervised Learning 计算机科学, 2022, 49(8): 12-25. https://doi.org/10.11896/jsjkx.210700111 |
| [2] | 杜丽君, 唐玺璐, 周娇, 陈玉兰, 程建. 基于注意力机制和多任务学习的阿尔茨海默症分类 Alzheimer's Disease Classification Method Based on Attention Mechanism and Multi-task Learning 计算机科学, 2022, 49(6A): 60-65. https://doi.org/10.11896/jsjkx.201200072 |
| [3] | 杨健楠, 张帆. 一种结合双注意力机制和层次网络结构的细碎农作物分类方法 Classification Method for Small Crops Combining Dual Attention Mechanisms and Hierarchical Network Structure 计算机科学, 2022, 49(6A): 353-357. https://doi.org/10.11896/jsjkx.210200169 |
| [4] | 郝强, 李杰, 张曼, 王路. 基于改进YOLOv3的空间非合作目标部件识别算法 Spatial Non-cooperative Target Components Recognition Algorithm Based on Improved YOLOv3 计算机科学, 2022, 49(6A): 358-362. https://doi.org/10.11896/jsjkx.210700048 |
| [5] | 朱旭东, 熊贇. 基于样本分布损失的图像多标签分类研究 Study on Multi-label Image Classification Based on Sample Distribution Loss 计算机科学, 2022, 49(6): 210-216. https://doi.org/10.11896/jsjkx.210300267 |
| [6] | 彭云聪, 秦小林, 张力戈, 顾勇翔. 面向图像分类的小样本学习算法综述 Survey on Few-shot Learning Algorithms for Image Classification 计算机科学, 2022, 49(5): 1-9. https://doi.org/10.11896/jsjkx.210500128 |
| [7] | 魏勤, 李瑛娇, 娄平, 严俊伟, 胡辑伟. 基于边云协同的人脸识别方法研究 Face Recognition Method Based on Edge-Cloud Collaboration 计算机科学, 2022, 49(5): 71-77. https://doi.org/10.11896/jsjkx.210300222 |
| [8] | 张文轩, 吴秦. 基于多分支注意力增强的细粒度图像分类 Fine-grained Image Classification Based on Multi-branch Attention-augmentation 计算机科学, 2022, 49(5): 105-112. https://doi.org/10.11896/jsjkx.210100108 |
| [9] | 许华杰, 陈育, 杨洋, 秦远卓. 基于混合样本自动数据增强技术的半监督学习方法 Semi-supervised Learning Method Based on Automated Mixed Sample Data Augmentation Techniques 计算机科学, 2022, 49(3): 288-293. https://doi.org/10.11896/jsjkx.210100156 |
| [10] | 董琳, 黄丽清, 叶锋, 黄添强, 翁彬, 徐超. 人脸伪造检测泛化性方法综述 Survey on Generalization Methods of Face Forgery Detection 计算机科学, 2022, 49(2): 12-30. https://doi.org/10.11896/jsjkx.210900146 |
| [11] | 苗启广, 辛文天, 刘如意, 谢琨, 王泉, 杨宗凯. 面向智慧教育行为分析的图卷积骨架动作识别方法 Graph Convolutional Skeleton-based Action Recognition Method for Intelligent Behavior Analysis 计算机科学, 2022, 49(2): 156-161. https://doi.org/10.11896/jsjkx.220100061 |
| [12] | 田天祎, 孙福明. 基于域自适应的肿瘤识别模型 Tumor Recognition Method Based on Domain Adaptive Algorithm 计算机科学, 2022, 49(12): 250-256. https://doi.org/10.11896/jsjkx.220600008 |
| [13] | 阿布都克力木·阿布力孜, 张雨宁, 阿力木江·亚森, 郭文强, 哈里旦木·阿布都克里木. 预训练语言模型的扩展模型研究综述 Survey of Research on Extended Models of Pre-trained Language Models 计算机科学, 2022, 49(11A): 210800125-12. https://doi.org/10.11896/jsjkx.210800125 |
| [14] | 张喜科, 马志庆, 赵文华, 崔冬梅. 基于卷积神经网络的乳腺癌组织病理学图像分类研究综述 Review on Classification of Breast Cancer Histopathological Images Based on Convolutional Neural Network 计算机科学, 2022, 49(11A): 210800232-9. https://doi.org/10.11896/jsjkx.210800232 |
| [15] | 原慧琳, 刘军涛, 黄碧, 韩真, 冯宠. 基于改进EfficientNetV2的有害垃圾图像分类方法 Classification Method of Harmful Garbage Images Based on Improved EfficientNetV2 计算机科学, 2022, 49(11A): 211100100-5. https://doi.org/10.11896/jsjkx.211100100 |
|
||
首页