基于分组残差结构的轻量级卷积神经网络设计

李鹏; 蒋品群; 曾上游; 夏海英; 廖志贤; 范瑞

Chinese

您当前的位置：

首页 >

文章列表页 >

基于分组残差结构的轻量级卷积神经网络设计

更新时间：2026-09-04

- 基于分组残差结构的轻量级卷积神经网络设计
- Design of lightweight convolution neural network based on group residual structure
- 微电子学与计算机 2019年36卷第7期页码：43-47
- 作者机构：
  
  广西师范大学机构电子工程学院,广西,桂林,541004
- 作者简介：
- 基金信息：
  
  国家自然科学基金 11465004国家自然科学基金 61762014桂林市科学研究与技术开发计划项目 20170113-4
- DOI：
  中图分类号：
- 纸质出版：2019
- 稿件说明：
移动端阅览
李鹏, 蒋品群, 曾上游, 夏海英, 廖志贤, 范瑞. 基于分组残差结构的轻量级卷积神经网络设计[J]. 微电子学与计算机, 2019, 36(7): 43-47.

LI Peng, JIANG Pin-qun, ZENG Shang-you, et al. Design of lightweight convolution neural network based on group residual structure[J]. 2019, 36(7): 43-47.
李鹏, 蒋品群, 曾上游, 夏海英, 廖志贤, 范瑞. 基于分组残差结构的轻量级卷积神经网络设计[J]. 微电子学与计算机, 2019, 36(7): 43-47. DOI：

LI Peng, JIANG Pin-qun, ZENG Shang-you, et al. Design of lightweight convolution neural network based on group residual structure[J]. 2019, 36(7): 43-47. DOI：

摘要

针对传统深度卷积神经网络参数数量过多

很难在移动设备上应用的问题

提出基于分组残差结构的轻量级卷积神经网络架构GResNets.利用三个卷积层的瓶颈结构将上层输出特征图分为数量相等的四组

根据组内的瓶颈模块加入恒等映射的方式和组外相邻模块是否加入残差学习

设计了三种轻量级卷积神经网络架构.试验阶段

在Caltech-256

Food-101和GTSRB图像分类数据集上评测了三种网络架构的性能.实验结果表明

与传统深度卷积神经网络相比

GResNets能在网络参数较少的情况下

具有同样、甚至更优越的分类性能

适合在移动设备上应用.

Abstract

In order to solve the problem that the parameters of traditional deep convolution neural network are too large to be used in mobile devices

a lightweight convolution neural network architecture GResNets based on group residual structure is proposed. Using the bottleneck structure with three convolution layers

the output feature maps of the previous layer are divided into four equal groups. Three lightweight convolution neural network architectures are designed according to the way of adding identical mapping to the bottleneck module in the group and whether or not the adjacent module outside the group joins residual learning. In the experimental stage

the performance of the three network architectures was evaluated on three image recognition datasets

Caltech-256

Food-101 and GTSRB. The experimental results show that GResNets has the same or even better classification performance than traditional deep convolution neural networks under the condition of fewer network parameters

and is suitable for mobile devices.

关键词

Keywords

references

PANDEY P, DEEPTHI A, MANDAL B, et al. FoodNet: recognizing foods using ensemble of deep networks[J]. IEEE Signal Processing Letters, 2017, 24(12): 1758-1762. DOI: 1758

KRIZHEVSKY A, SUTSKEVER I, HINTON G E. Imagenet classification with deep convolutional neural networks[C]//Advances in neural information processing systems. [s.l.]2012: 1097-1105.

JIA Y, SHELHAMER E, DONAHUE J, et al. Caffe: Convolutional architecture for fast feature embedding[C]//Proceedings of the 22nd ACM international conference on Multimedia. ACM, 2014: 675-678.

XIE S, GIRSHICK R, DOLLAR P, et al. Aggregated residual transformations for deep neural networks[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). USA, Honolulu, IEEE, 2017: 5987-5995.

SERMANET P, LECUN Y. Traffic sign recognition with multi-scale convolutional networks[C]//Neural Networks (IJCNN), The 2011 International Joint Conference on. USA, San Jose, IEEE, 2011: 2809-2813.

GONZALEZ-REYNA S E, AVINA-CERVANTES J G, LEDESMA-OROZCO S E, et al. Traffic sign recognition based on linear discriminant analysis[C]//Mexican International Conference on Artificial Intelligence. Springer, Berlin, Heidelberg, 2013: 185-193.

HERAVI E J, AGHDAM H H, PUIG D. An optimized convolutional neural network with bottleneck and spatial pyramid pooling layers for classification of foods[J]. Pattern Recognition Letters, 2018(105): 50-58. DOI: 50

IANDOLA F N, HAN S, MOSKEWICZ M W, et al. Squeezenet: Alexnet-level accuracy with 50x fewer parameters and < 0.5 mb model size[J]. arXiv preprint arXiv:1602.07360, 2016.

杨远飞, 曾上游, 冯燕燕, 等.基于并行和切片的深度卷积网络设计研究[J].微电子学与计算机, 2018, 35(3): 120-124. DOI: 120

CHOLLET F. Xception: Deep learning with depthwise separable convolutions[J]. arXiv preprint arXiv: 1610.02357, 2017.

HOWARD A G, ZHU M, CHEN B, et al. Mobilenets: Efficient convolutional neural networks for mobile vision applications[J]. arXiv preprint arXiv:1704.04861, 2017.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

面向处理器分支预测单元的Spectre攻击轻量化防御方法

AlN/GaN混合缓冲层双异质结p-GaN HEMT直流特性仿真

基于区块链的AI生成内容溯源框架设计研究

大模型发展综述：从BERT到Deepseek的技术演进

基于改进RT-DETR的航拍小目标检测算法