基于奇偶交叉卷积的轻量级图像语义分割网络

doi:10.3969/j.issn.0255-8297.2022.03.008

应用科学学报 ›› 2022, Vol. 40 ›› Issue (3): 448-456.doi: 10.3969/j.issn.0255-8297.2022.03.008

基于奇偶交叉卷积的轻量级图像语义分割网络

栗风永¹, 叶彬¹, 秦川²

1. 上海电力大学计算机科学与技术学院, 上海 201306;
2. 上海理工大学光电信息与计算机工程学院, 上海 200093

收稿日期:2021-01-26 发布日期:2022-05-25
通信作者: 栗风永,副教授,研究方向为多媒体信息安全机器学习。E-mail:fyli@shiep.edu.cn E-mail:fyli@shiep.edu.cn
基金资助:
国家自然科学基金(No.U1736120);上海市自然科学基金(No.20ZR1421600)资助

Lightweight Image Semantic Segmentation Network Based on Parity Cross Convolution

LI Fengyong¹, YE Bin¹, QIN Chuan²

1. College of Computer Science and Technology, Shanghai University of Electric Power, Shanghai 201306, China;
2. School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China

Received:2021-01-26 Published:2022-05-25

摘要/Abstract

摘要： 传统轻量级图像语义分割网络中的跨步下采样卷积导致感受野呈现刺状分布，使得像素点利用率出现系统偏差，影响分割精度的提升。为此，针对传统的轻量级图像语义分割网络设计一种奇偶交叉卷积的下采样模块，在跨步奇数卷积模块前增加单步偶数卷积模块，在一定程度上缓解了刺状分布带来的不良影响，消除了分割网络中不同空间位置上像素利用率的偏差，最终提高了模型对像素点的分割精度。通过7种不同轻量级图像语义分割网络的对比可以看出，所提模型可以明显消除刺状分布，使分割网络的精度进一步提高，同时也证明了该模型适用于不同的轻量级网络，具有普适性。

关键词: 图像语义分割, 卷积神经网络, 轻量级网络, 奇偶交叉卷积

Abstract: The multi-step down-sampling convolution in traditional lightweight image semantic segmentation networks easily causes a thorn-like distribution in receptive fields. This will introduce systematic deviations in pixel utilization and finally affect the improvement of segmentation accuracy. Regarding this problem, a down-sampling module of parity cross convolution is designed for traditional lightweight image semantic segmentation networks. In the scheme, an even-convolution module is added before the strided odd-numbered convolution module, in order to alleviate the negative effects caused by spur distribution. Accordingly, the deviation of pixel utilization at different spatial positions in the segmentation network can be eliminated effectively, and the improvement of pixel segmentation accuracy can be finally achieved. Experimental results demonstrate that through the comparison of seven different lightweight image semantic segmentation networks, the proposed model can obviously eliminate the thorn-like distribution and improve the accuracy of the segmentation network. Also, the proposed model performs excellent adaptability to different lightweight networks.

Key words: image semantic segmentation, convolutional neural network (CNN), lightweight network, parity cross convolution

中图分类号:

TP391

栗风永, 叶彬, 秦川. 基于奇偶交叉卷积的轻量级图像语义分割网络[J]. 应用科学学报, 2022, 40(3): 448-456.

LI Fengyong, YE Bin, QIN Chuan. Lightweight Image Semantic Segmentation Network Based on Parity Cross Convolution[J]. Journal of Applied Sciences, 2022, 40(3): 448-456.

参考文献

[1] 李欣,唐文莉,杨博.利用深度残差网络的高分遥感影像语义分割[J].应用科学学报, 2019, 37(2):282-290. Li X, Tang W L, Yang B. Semantic segmentation of high-resolution remote sensing image based on deep residual network[J]. Journal of Applied Sciences, 2019, 37(2):282-290.(in Chinese)
[2] Luo W, Li Y, Urtasun R, et al. Understanding the effective receptive field in deep convolutional neural networks[C]//Proceedings of the 30th International Conference on Neural Information Processing Systems, 2016:4905-4913.
[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 and Machine Intelligence, 2017, 40(4):834-848.
[4] Wang P, Chen P, Yuan Y, et al. Understanding convolution for semantic segmentation[C]//2018 IEEE Winter Conference on Applications of Computer Vision (WACV). IEEE, 2018:1451-1460.
[5] Howard A, Sandler M, Chu G, et al. Searching for mobilenetv3[C]//Proceedings of the IEEE International Conference on Computer Vision, 2019:1314-1324.
[6] Yu C, Wang J, Peng C, et al. Bisenet:bilateral segmentation network for real-time semantic segmentation[C]//Proceedings of the European Conference on Computer Vision (ECCV), 2018:325-341.
[7] 李宝奇,贺昱曜,何灵蛟,等.基于全卷积神经网络的非对称并行语义分割模型[J].电子学报, 2019, 47(5):1058-1064. Li B Q, He Y Y, He L J, et al. Asymmetric parallel semantic segmentation model based on full convolutional neural network[J]. Acta Electronica Sinica, 2019, 47(5):1058-1064.(in Chinese)
[8] 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. Han H H, Li W T, Wang J P, et al. Semantic segmentation of encoder-decoder structure[J]. Journal of Image and Graphics, 2020, 25(2):255-266.(in Chinese)
[9] Poudel R P K, Liwicki S, Cipolla R. Fast-SCNN:fast semantic segmentation network[DB/OL]. 2019[2019-02-12]. https://arxiv.org/abs/1902.04502.
[10] Yu C, Gao C, Wang J, et al. BiSeNet V2:bilateral network with guided aggregation for realtime semantic segmentation[DB/OL]. 2020[2020-04-05]. https://arxiv.org/abs/2004.02147.
[11] Poudel R P K, Bonde U, Liwicki S, et al. Contextnet:exploring context and detail for semantic segmentation in real-time[DB/OL]. 2018[2018-11-05]. https://arxiv.org/abs/1805.04554.
[12] Li G, Yun I, Kim J, et al. DABNet:depth-wise asymmetric bottleneck for real-time semantic segmentation[DB/OL]. 2019[2019-10-01]. https://arxiv.org/abs/1907.11357.
[13] Li X, Zhou Y, Pan Z, et al. Partial order pruning:for best speed/accuracy trade-off in neural architecture search[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2019:9145-9153.
[14] Liu M, Yin H. Feature pyramid encoding network for real-time semantic segmentation[DB/OL]. 2019[2019-09-18]. https://arxiv.org/abs/1909.08599.

[1]	朱丽, 杨青, 吴涛, 李晨, 李铭. 基于CNN和Bi-LSTM的脑电波情感分析[J]. 应用科学学报, 2022, 40(1): 1-12.
[2]	周传华, 徐文倩, 朱俊杰. 基于代价敏感卷积神经网络的集成分类算法[J]. 应用科学学报, 2022, 40(1): 69-79.
[3]	李文举, 张耀星, 陈慧玲, 李培刚, 沙利业. 基于TSCD模型的轨道板裂缝检测方法[J]. 应用科学学报, 2022, 40(1): 155-166.
[4]	郑长亮, 庞明. 基于卷积神经网络的时空权重姿态运动特征提取算法[J]. 应用科学学报, 2021, 39(4): 594-604.
[5]	查玉坤, 张其林, 赵永标, 杭波. 基于三维卷积和CLSTM神经网络的水产养殖溶解氧预测[J]. 应用科学学报, 2021, 39(4): 615-626.
[6]	李文举, 何茂贤, 张耀星, 陈慧玲, 李培刚. 基于卷积神经网络和投票机制的轨道板裂缝检测[J]. 应用科学学报, 2021, 39(4): 627-640.
[7]	郝琰, 石慧宇, 霍首君, 韩丹, 曹锐. 基于脑电信号深度学习的情感分类[J]. 应用科学学报, 2021, 39(3): 347-346.
[8]	崔鹏涛, 张倩, 刘敬怀, 周超, 王斌, 司文. 基于FSCD-CNN的深度图像快速帧内预测模式选择算法[J]. 应用科学学报, 2021, 39(3): 433-432.
[9]	李磊, 张青苗, 赵军辉, 聂逸文. 基于改进CNN-LSTM组合模型的分时段短时交通流预测[J]. 应用科学学报, 2021, 39(2): 185-198.
[10]	张涵, 秦昆, 毕奇, 张晔, 许凯. 注意力引导的三维卷积网络用于遥感场景变化检测[J]. 应用科学学报, 2021, 39(2): 272-280.
[11]	刘之瑜, 张淑芬, 刘洋, 罗长银, 李敏. 基于图像梯度的数据增广方法[J]. 应用科学学报, 2021, 39(2): 302-311.
[12]	王孟轩, 张胜, 王月, 雷霆, 杜渂. 改进的CRNN模型在警情文本分类中的研究与应用[J]. 应用科学学报, 2020, 38(3): 388-400.
[13]	马鑫, 吴云, 鹿泽光. 基于混合神经网络的协同过滤推荐模型[J]. 应用科学学报, 2020, 38(3): 478-487.
[14]	刘伟, 章琬苓, 项世军. 基于LBP-MDCT和CNN的人脸活体检测算法[J]. 应用科学学报, 2019, 37(5): 609-617.
[15]	王灿军, 廖鑫, 陈嘉欣, 秦拯, 刘绪崇. 基于卷积神经网络的面部图像修饰检测[J]. 应用科学学报, 2019, 37(5): 618-630.

基于奇偶交叉卷积的轻量级图像语义分割网络

Lightweight Image Semantic Segmentation Network Based on Parity Cross Convolution

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价