计算机工程与应用 ›› 2022, Vol. 58 ›› Issue (5): 85-93.DOI: 10.3778/j.issn.1002-8331.2101-0020

• 理论与研发 • 上一篇    下一篇

应用空洞卷积的神经网络轻量化方法#br#

马利,刘新宇,李皓宇,段苛苛,牛斌   

  1. 辽宁大学 信息学院,沈阳 110036
  • 出版日期:2022-03-01 发布日期:2022-03-01

Neural Network Lightweight Method with Dilated Convolution

MA Li, LIU Xinyu, LI Haoyu, DUAN Keke, NIU Bin   

  1. School of Information, Liaoning University, Shenyang 110036, China
  • Online:2022-03-01 Published:2022-03-01

摘要: 为了深度卷积神经网络能够更好地应用于边缘设备,减少深度神经网络的模型参数量,降低网络复杂度,对于轻量化卷积神经网络的研究日益增多。将常用于多尺度融合的卷积——空洞卷积首次应用于神经网络轻量化领域研究。利用空洞卷积采样与扩大感受野的结构特性,实现了模型参数量的减少与计算复杂度的降低。将空洞卷积单纯作用于残差网络结构,达到轻量化目的,并进一步与逐点卷积结合,提高轻量化效果,形成改进型空洞卷积轻量化方法。为减少准确率衰减,将改进型空洞卷积与普通卷积相融合,提出一种融合型空洞卷积轻量化方法。实验结果表明,改进型空洞卷积轻量化方法具有最显著的轻量化效果,融合型空洞卷积轻量化方法使模型参数量减少同时具有最佳速度与精度的权衡。

关键词: 卷积神经网络, 轻量化, 空洞卷积, 融合, 精度与速度权衡

Abstract: In order to apply deep convolutional neural networks better to edge devices, reduce the amount of deep neural networks’ model parameters, and reduce network complexity, research on lightweight convolutional neural networks is increasing. In this article, some researches are first done in the field of lightweight neural networks about dilated convolution which is commonly used in multi-scale fusion. Using the structural characteristics of dilated convolution sampling and expanding the receptive field, the reduction of model parameters and the reduction of computational complexity are realized. Firstly, the dilated convolution is purely applied to the residual network structure to light convolutional neural networks, and further combined dilated convolution with the pointwise convolution to improve the lightweight effect, it proposes improved dilated convolution lightweight method. In order to reduce the attenuation of accuracy, it combines improved dilated convolution and ordinary convolution, and proposes a lightweight method of fusing dilated convolution. Experimental results show that the improved dilated  convolution lightweight method has the most significant lightweight effect, the fusing dilated convolution lightweight method not only reduces the amount of model parameters, but also has the best speed and accuracy trade-off.

Key words: convolutional neural networks, lightweight method, dilated convolution, fusion, trade-off between accuracy and speed