[1]陈莉君,李 卓.基于深度神经压缩的 YOLO 优化[J].计算机技术与发展,2019,29(12):72-75.[doi:10. 3969 / j. issn. 1673-629X. 2019. 12. 013]
　CHEN Li-jun,LI Zhuo.YOLO Optimization Based on Deep Neural Compression[J].,2019,29(12):72-75.[doi:10. 3969 / j. issn. 1673-629X. 2019. 12. 013]
点击复制

基于深度神经压缩的 YOLO 优化()
分享到：

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

卷:
29
期数:
2019年12期
页码:
72-75
栏目:
智能、算法、系统工程
出版日期:
2019-12-10

文章信息/Info

Title:
YOLO Optimization Based on Deep Neural Compression
文章编号:
1673-629X(2019)12-0072-04
作者:
陈莉君; 李 卓
西安邮电大学 计算机学院,陕西 西安 710100
Author(s):
CHEN Li-jun; LI Zhuo
School of Computer,Xi’an University of Posts &Telecommunications,Xi’an 710100,China
关键词:
模型压缩; 深度学习; 目标检测; 权重量化
Keywords:
model compression; deep learning; object detection; weight quantification
分类号:
TP183
DOI:
10. 3969 / j. issn. 1673-629X. 2019. 12. 013
摘要:
从 AlphaGo 开始,深度学习逐渐进入业内研究者的视线。 深度学习成为热点的主要原因是由于近些年设备的计算力的增加,尤其是图形处理器对于浮点数运算的有力支持。 YOLO 在提出一种新的目标检测方法的同时,由于其过多的网络层数,带来对于存储空间巨大的需求。 因此需要对于模型进行压缩,减少对于存储空间的需求。 在传统压缩过程中单独使用剪枝或者量化方法,压缩后的模型依然存在一定的冗余。 因此提出了一个结合剪枝和量化的方法对于模型进行压缩。 文中针对在原始 YOLO 模型没有对于模型的测试方法以及对于模型稀疏度评估的手段进行优化。 在压缩的过程展示中,明确地标明每一层的稀疏度。 实验结果证明,YOLO 模型在 VOC2012 数据集条件下,在保持接近原始模型的精度情况下,压缩了 10 倍。
Abstract:
Beginning with AlphaGo,deep learning has gradually entered the eyes of industry researchers. The main reason for deep learning is the increase in the computing power of devices in recent years,especially the strong support of graphics processors for floating-point operations. While YOLO proposes a new object detection method,it has a huge demand for storage space due to its excessive number of network layers. Therefore,it is necessary to compress the model to reduce the need for storage space. In the traditional compression process,pruning or quantification methods are used separately,and the compressed model still has some redundancy. Therefore,a method combining pruning and quantification is proposed to compress the model. We optimize for the original YOLO model without the test method for the model and the means for model sparsity evaluation. In the process of compression,the sparsity of each layer is clearly indicated. The experiment shows that the YOLO model is compressed 10 times under the condition of VOC2012 data set while maintaining the accuracy close to the original model.

相似文献/References:

[1]陈强锐,谢世朋.基于深度学习的肺部肿瘤检测方法[J].计算机技术与发展,2018,28(04):201.[doi:10.3969/ j. issn.1673-629X.2018.04.043]
　CHEN Qiang-rui,XIE Shi-peng.Lung Cancer Detection Method Based on Deep Learning[J].,2018,28(12):201.[doi:10.3969/ j. issn.1673-629X.2018.04.043]
[2]施泽浩,赵启军.基于全卷积网络的目标检测算法[J].计算机技术与发展,2018,28(05):55.[doi:10.3969/j.issn.1673－629X.2018.05.013]
　SHI Ze-hao,ZHAO Qi-jun.Object Detection Algorithm Based on Fully Convolutional Neural Network[J].,2018,28(12):55.[doi:10.3969/j.issn.1673－629X.2018.05.013]
[3]黄法秀,张世杰,吴志红,等.数据增广下的人脸识别研究[J].计算机技术与发展,2020,30(03):67.[doi:10. 3969 / j. issn. 1673-629X. 2020. 03. 013]
　HUANG Fa-xiu,ZHANG Shi-jie,WU Zhi-hong,et al.Research on Face Recognition Based on Data Augmentation[J].,2020,30(12):67.[doi:10. 3969 / j. issn. 1673-629X. 2020. 03. 013]
[4]陈浩翔,蔡建明,刘铿然,等. 手写数字深度特征学习与识别[J].计算机技术与发展,2016,26(07):19.
　CHEN Hao-xiang,CAI Jian-ming,LIU Keng-ran,et al. Deep Learning and Recognition of Handwritten Numeral Features[J].,2016,26(12):19.
[5]高翔,陈志,岳文静,等.基于视频场景深度学习的人物语义识别模型[J].计算机技术与发展,2018,28(06):53.[doi:10.3969/ j. issn.1673-629X.2018.06.012]
　GAO Xiang,CHEN Zhi,YUE Wen-jing,et al.Human Semantic Recognition Model Based on Video Scene Deep Learning[J].,2018,28(12):53.[doi:10.3969/ j. issn.1673-629X.2018.06.012]
[6]贺飞翔,赵启军. 基于深度学习的头部姿态估计[J].计算机技术与发展,2016,26(11):1.
　HE Fei-xiang,ZHAO Qi-jun. Head Pose Estimation Based on Deep Learning[J].,2016,26(12):1.
[7]徐 融,邱晓晖.一种改进的 YOLO V3 目标检测方法[J].计算机技术与发展,2020,30(07):30.[doi:10. 3969 / j. issn. 1673-629X. 2020. 07. 007]
　XU Rong,QIU Xiao-hui.An Improved YOLO V3 Object Detection[J].,2020,30(12):30.[doi:10. 3969 / j. issn. 1673-629X. 2020. 07. 007]
[8]曾志平[] [],萧海东[],张新鹏[]. 基于DBN的金融时序数据建模与决策[J].计算机技术与发展,2017,27(04):1.
　ZENG Zhi-ping[] [],XIAO Hai-dong[],ZHANG Xin-peng[]. Modeling and Decision-making of Financial Time Series Data with DBN[J].,2017,27(12):1.
[9]李全兵,文 钊*,田艳梅*,等.基于 WGAN 的音频关键词识别研究[J].计算机技术与发展,2021,31(08):26.[doi:10. 3969 / j. issn. 1673-629X. 2021. 08. 005]
　LI Quan-bing,WEN Zhao *,TIAN Yan-mei *,et al.Research on Audio Keywords Recognition Based on WassersteinGenerative Adversarial Network[J].,2021,31(12):26.[doi:10. 3969 / j. issn. 1673-629X. 2021. 08. 005]
[10]李宏林. 分析式纹理合成技术及其在深度学习的应用[J].计算机技术与发展,2017,27(11):7.
　LI Hong-lin. Analyzed Texture-synthesis Techniques and Their Applications in Deep Learning[J].,2017,27(12):7.
[11]张佳钰,寇金桥,刘宁钟.基于滤波器分布拟合的神经网络剪枝算法[J].计算机技术与发展,2022,32(12):136.[doi:10. 3969 / j. issn. 1673-629X. 2022. 12. 021]
　ZHANG Jia-yu,KOU Jin-qiao,LIU Ning-zhong.Deep Convolutional Neural Networks Pruning Algorithm Based on Filter Pruning via Distribution Fitting[J].,2022,32(12):136.[doi:10. 3969 / j. issn. 1673-629X. 2022. 12. 021]

更新日期/Last Update: 2019-12-10