改进DeepLabV3+的高效语义分割

计算机工程与科学 ›› 2022, Vol. 44 ›› Issue (04): 737-745.

改进DeepLabV3+的高效语义分割

马冬梅，李鹏辉，黄欣悦，张倩，杨鑫

（西北师范大学物理与电子工程学院，甘肃兰州 730070）

收稿日期:2020-09-25 修回日期:2020-11-02 接受日期:2022-04-25 出版日期:2022-04-25 发布日期:2022-04-20
基金资助:
国家自然科学基金（61961037）

Efficient semantic segmentation based on improved DeepLabV3+

MA Dong-mei,LI Peng-hui,HUANG Xin-yue,ZHANG Qian,YANG Xin

（School of Physics & Electronic Engineering,Northwest Normal University,Lanzhou 730070,China）

Received:2020-09-25 Revised:2020-11-02 Accepted:2022-04-25 Online:2022-04-25 Published:2022-04-20

摘要/Abstract

摘要： 针对目前高精度的语义分割模型普遍存在计算复杂度高、占用内存大，难以在硬件存储和计算力有限的嵌入式平台部署的问题，从网络的参数量、计算量和性能3个方面综合考虑，提出一种基于改进DeepLabV3+的高效语义分割模型。该模型以MobileNetV2为骨干网络，在空洞空间金字塔池化（ASPP）模块中并联混合带状池化（MSP），以获取密集的上下文信息；在解码部分引入有效通道注意力（ECA）模块，以恢复更清晰的目标边界；将深度可分离卷积应用到ASPP模块和解码器中用于压缩模型。在PASCAL VOC 2012数据集上的实验中，该模型的网络参数量为4.5×106，浮点计算量为11.13 GFLOPs，平均交并比为72.07%，在计算效率和分割精度之间达到了良好的均衡。

关键词: 语义分割, DeepLabV3+, 带状池化, 有效通道注意力, 深度可分离卷积

Abstract: The current high-precision semantic segmentation medel generally have the problems of high computational complexity and large memory usage, so it is difficult to deploy on embedded platforms with limited hardware storage and computing power. Aiming at the problem, an improved efficient semantic segmentation medel based on improved DeepLabV3+ is proposed by comprehensively considering three aspects of network parameters, calculation and performance. The model uses MobileNetV2 as the backbone network, and combines the mix strip pooling(MSP) in the atrous spatial pyramid pooling (ASPP) module to obtain dense context information. The effective channel attention (ECA) module is introduced in the decoder to restore a clearer target boundary. Depthwise separable convolution is applied to the ASPP module and decoder to compress the model. Experiment on the PASCAL VOC 2012 dataset show that the number of network parameters of the medel is 4.5×106, the number of floating point operations is 11.13 GFLOPs, and the mean intersection over union is 72.07%, which proves that the algorithm achieves the good balance between calculation efficiency and segmentation accuracy.

Key words: semantic segmentation, DeepLabV3+, strip pooling, efficient channel attention, depthwise separable convolution

马冬梅, 李鹏辉, 黄欣悦, 张倩, 杨鑫. 改进DeepLabV3+的高效语义分割[J]. 计算机工程与科学, 2022, 44(04): 737-745.

MA Dong-mei, LI Peng-hui, HUANG Xin-yue, ZHANG Qian, YANG Xin. Efficient semantic segmentation based on improved DeepLabV3+[J]. Computer Engineering & Science, 2022, 44(04): 737-745.

参考文献［17］

［1］	Long J,Shelhamer E,Darrell T.Fully convolutional networks for semantic segmentation［C］∥Proc of the 2015 IEEE Conference on Computer Vision and Pattern Recognition,2015:3431-3440.
［2］	Ronneberger O,Fischer P,Brox T.U-Net:Convolutional networks for biomedical image segmentation［C］∥Proc of the International Conference on Medical Image Computing and Computer Assisted Intervention,2015:234-241.
［3］	Chen L C,Papandreou G,Kokkinos I,et al.DeepLab:Semantic image segmentation with deep convolutional nets,atrous convolution,and fully connected CRFs［J］.IEEE Transactions on Pattern Analysis & Machine Intelligence,2016,40(4):834-848.
［4］	Chen L C,Papandreou G,Schroff F,et al.Rethinking atrous convolution for semantic image segmentation［C］∥Proc of the 2017 IEEE Conference on Computer Vision and Pattern Recognition,2017:1-4.
［5］	Chen L C,Zhu Y,Papandreou G,et al.Encoder-decoder with atrous separable convolution for semantic image segmentation［C］∥Proc of European Conference on Computer Vision,2018:801-818.
［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］	Sandler M,Howard A,Zhu M,et al.MobileNetV2:Inverted residuals and linear bottlenecks［C］∥Proc of the 2018 IEEE Conference on Computer Vision and Pattern Recognition,2018:4510-4520.
［8］	Zhang X,Zhou X,Lin M,et al.ShuffleNet:An extremely efficient convolutional neural network for mobile devices［C］∥Proc of the 2018 IEEE Conference on Computer Vision and Pattern Recognition,2018:6848-6856.
［9］	Chollet F.Xception:Deep learning with depthwise separable convolutions［C］∥Proc of the 2017 IEEE Conference on Computer Vision and Pattern Recognition,2017:1800-1807.
［10］	Yang T,Howard A,Chen B,et al. NetAdapt:Platform- aware neural network adaptation for mobile applications［C］∥Proc of European Conference on Computer Vision,2018:289-304.
［11］	Jacob B,Kligys S,Chen B,et al.Quantization and training of neural networks for efficient integer-arithmetic-only infer- ence［C］∥Proc of the 2018 IEEE Conference on Computer Vision and Pattern Recognition,2018:2704-2713.
［12］	Hinton G,Vinyals O,Dean J.Distilling the knowledge in a neural network［J］.arXiv:1503.02531,2015.
［13］	Mehta S,Rastegari M,Caspi A,et al.ESPNet:Efficient spatial pyramid of dilated convolutions for semantic segmentation［C］∥Proc of the 2018 IEEE Conference on Computer Vision and Pattern Recognition,2018:561-580.
［14］	Li Yu-yan,Zhang Yin-hui, He Zi-fen.Semantic segmentation of tennis scene based on series atrous convolution neural network ［J］.Journal of Computer-Aided Design & Computer Graphics,2020,32(4):606-615.(in Chinese)
［15］	Kipf T,Welling M.Semi-supervised classification with graph convolutional networks［J］.arXiv:1609.02907,2016.
［16］	Han Hui-hui, Li Wei-tao,Wang Jian-ping,et al.Semantic segmentation of encoder-decoder structure［J］． Journal of Image and Graphics，2020,25(2):255-266.(in Chinese)
［17］	Zhou Deng-wen,Tian Jin-yue,Ma Lu-yao,et al.Lightweight image semantic segmentation based on multi-level feature cascaded network ［J］.Journal of Zhejiang University (Engineering Science),2020,54(8):1516-1524.(in Chinese)
［18］	Szegedy C,Vanhoucke V,Ioffe S,et al.Rethinking the inception architecture for computer vision［C］∥Proc of the 2016 IEEE Conference on Computer Vision and Pattern Recognition,2016:2818-2826.
［19］	Zhao H,Shi J,Qi X,et al.Pyramid scene parsing network［C］∥Proc of the 2017 IEEE Conference on Computer Vision and Pattern Recognition,2017:6230-6239.
［20］	Hou Q B, Zhang L, Cheng M M, et al.Strip pooling:Rethinking spatial pooling for scene parsing［C］∥Proc of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition,2020:4003-4012.
［21］	Hu J,Shen L,Sun G.Squeeze-and-excitation networks［C］∥Proc of the 2018 IEEE Conference on Computer Vision and Pattern Recognition,2018:7132-7141.
［22］	Zhao H,Zhang Y,Liu S,et al.PSANet:Point-wise spatial attention network for scene parsing［C］∥Proc of European Conference on Computer Vision,2018:270-286.
［23］	Fu J,Liu J,Tian H,et al.Dual attention network for scene segmentation［C］∥Proc of the 2019 IEEE Conference on Computer Vision and Pattern Recognition,2019:3146-3154.
［24］	Huang Z, Wang X, Huang L, et al.CCNET:Criss-cross attention for semantic segmentation［C］∥Proc of the IEEE International Conference on Computer Vision,2019:603-612.
［25］	Yu C,Wang J,Peng C,et al.BiSeNet:Bilateral segmentation network for real-time semantic segmentation［C］∥Proc of European Conference on Computer Vision,2018:334-349.
［26］	Wang Q L, Wu B G, Zhu P F, et al.ECA-Net:Efficient channel attention for deep convolutional neural networks［C］∥Proc of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition,2020:1534-11542.
［27］	Ruderman A,Rabinowitz N C,Morcos A S,et al.Pooling is neither necessary nor sufficient for appropriate deformation stability in CNNs［J］.arXiv:1804.04438,2018.
［28］	He K M, Girshick R,Dollár P.Rethinking ImageNet pre- training［C］∥Proc of the 2019 IEEE/CVF International Conference on Computer Vision,2019:4918-4927.
	附中文参考文献：
［14］	李逾严,张印辉,何自芬.基于串联空洞卷积神经网络的网球场景语义分割［J］.计算机辅助设计与图形学学报,2020,32(4):606-615.
［16］	韩慧慧,李帷韬,王建平,等.编码—解码结构的语义分割［J］.中国图象图形学报,2020,25(2):255-266.
［17］	周登文,田金月,马路遥,等.基于多级特征并联的轻量级图像语义分割［J］.浙江大学学报(工学版),2020,54(8):1516-1524.