用于宫颈细胞分类的轻量级网络ICA-Res2Net

doi:10.3778/j.issn.1002-8331.2209-0014

摘要/Abstract

摘要： 针对目前宫颈细胞分类准确率不高以及实时性差的问题，提出改进的协调注意力模块（improved coordinate attention，ICA），并结合新型残差结构Res2Net以及空间金字塔池化层，设计一种轻量级深度卷积神经网络ICA-Res2Net。利用Res2Net网络特征子块间的交叉卷积，提取特征层中更细粒度的信息；采用空间金字塔池化提取局部区域特征，从而在不增加训练参数量的同时有效提取特征；进一步引入改进的轻量级注意力模块，通过横向池化、纵向池化等操作，给予特征层各像素点不同的加权值，强化重要细节特征，帮助网络定位感兴趣对象。此外，为有效防止深度网络的退化，提出的ICA-Res2Net网络保留了残差网络的跳跃连接设计；并联合Softmax损失函数和中心损失函数对网络参数进行训练，提高其分类准确率。利用提出的轻量级网络对SIPaKMeD公开数据集中的宫颈细胞图像进行分类，测试集的分类准确率达到98.65%，且网络的训练参数比ResNet50、DenseNet121等经典网络更少，显著提升宫颈细胞图像的分类效率。

关键词: 宫颈细胞, 深度卷积神经网络, 图像分类, 多尺度特征, 注意力机制

Abstract: To solve the problems of low accuracy and poor real-time property for cervical cell classification, this paper proposes an improved coordinate attention (ICA) module, and designs a lightweight deep convolution neural network ICA-Res2Net by combining with the new residual structure Res2Net and spatial pyramid pooling layer. Firstly, cross-convolution between feature sub-blocks of Res2Net network is adopted to extract finer granularity information in feature layer. Then, the spatial pyramid pooling is used to extract local regional features, thus the features can be effectively extracted without increasing the number of training parameters. The improved lightweight attention module is further introduced to weight each pixel in the feature layer through operations such as horizontal pooling and vertical pooling, so as to strengthen the important detailed features and help the network locating the objects of interest. In addition, in order to effectively prevent the degradation of deep network, the proposed ICA-Res2Net network retains the design of skip connection in residual network; and the network parameters are trained by combined the Softmax loss function with the center loss function to improve the classification accuracy. Applying the lightweight network proposed in this paper to classify cervical cell images in the SIPaKMeD public dataset, the test classification accuracy can reach 98.65%, and the training parameters of the network are much fewer than those of the classic networks such as ResNet50 and DenseNet121, which significantly improves the classification efficiency of cervical cell images.

Key words: cervical cell, deep convolutional neural network, image classification, multi-scale characteristics, attention mechanism

张鹏, 谢莉, 杨海麟. 用于宫颈细胞分类的轻量级网络ICA-Res2Net[J]. 计算机工程与应用, 2024, 60(3): 187-195.

ZHANG Peng, XIE Li, YANG Hailin. Lightweight Network ICA-Res2Net for Cervical Cell Classification[J]. Computer Engineering and Applications, 2024, 60(3): 187-195.

参考文献

[1] SUNG H, FERLAY J, SIEGEL R L, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J]. CA: A Cancer Journal for Clinicians, 2021, 71(3): 209-249.
[2] ZHANG S, XU H, ZHANG L, et al. Cervical cancer: epidemiology, risk factors and screening[J]. Chinese Journal of Cancer Research, 2020, 32(6): 720-728.
[3] 董娜, 赵丽, 常建芳, 等. 宫颈细胞图像的特征选择与分类识别算法研究[J]. 湖南大学学报 (自然科学版), 2019, 46(12): 1-8.
DONG N, ZHAO L, CHANG J F, et al. Research on feature selection and classification recognition algorithm of cervical cell image[J]. Journal of Hunan University (Natural Sciences), 2019, 46(12): 1-8.
[4] 赵理莉, 孙燎原, 殷建平, 等. 结合层次法与主成分分析特征变换的宫颈细胞识别[J]. 国防科技大学学报, 2017, 39(6): 45-50.
ZHAO L L, SUN L Y, YIN J P, et al. Cervical cell recognition based on hierarchical method and principal component analysis feature transformation[J]. Journal of National University of Defense Technology, 2017, 39(6): 45-50.
[5] MARIARPUTHAM E J, STEPHEN A. Nominated texture based cervical cancer classification[J]. Computational and Mathematical Methods in Medicine, 2015(2): 586928.
[6] 杨培伟, 周余红, 邢岗, 等. 卷积神经网络在生物医学图像上的应用进展[J]. 计算机工程与应用, 2021, 57(7): 44-58.
YANG P W, ZHOU Y H, XING G, et al. Applications of convolutional neural network in biomedical image[J]. Computer Engineering and Applications, 2021, 57(7): 44-58.
[7] 张梦倩, 张莉. 粗-细两阶段卷积神经网络算法[J]. 计算机科学与探索, 2021, 15(8): 1501-1510.
ZHANG M Q, ZHANG L. Coarse-to-fine two-stage convolutional neural network algorithm[J]. Journal of Frontiers of Computer Science and Technology, 2021, 15(8): 1501-1510.
[8] 赵越, 曾立波, 吴琼水. 卷积神经网络的宫颈细胞图像分类[J]. 计算机辅助设计与图形学学报, 2018, 30(11): 2049-2054.
ZHAO Y, ZENG L B, WU Q S. Classification of cervical cells based on convolution neural network[J]. Journal of Computer-Aided Design & Computer Graphics, 2018, 30(11): 2049-2054.
[9] PLISSITI M E, DIMIKRAKOPOULOS P, SFIKAS G, et al. SIPAKMED: a new dataset for feature and image based classification of normal and pathological cervical cells in Pap smear images[C]//2018 25th IEEE International Conference on Image Processing (ICIP), Athens, Greece, Oct 7-10, 2018. Piscataway: IEEE, 2018: 3144-3148.
[10] GHONEIM A, MUHAMMAD G, HOSSAIN M S. Cervical cancer classification using convolutional neural networks and extreme learning machines[J]. Future Generation Computer Systems, 2020, 102: 643-649.
[11] SHI J, WANG R, ZHENG Y, et al. Cervical cell classification with graph convolutional network[J]. Computer Methods and Programs in Biomedicine, 2021, 198: 105807.
[12] 周涛, 刘赟璨, 陆惠玲, 等. ResNet及其在医学图像处理领域的应用: 研究进展与挑战[J]. 电子与信息学报, 2022, 44(1): 149-167.
ZHOU T, LIU Y C, LU H L, et al. ResNet and its application to medical image processing: research progress and challenges[J]. Journal of Electronics & Information Technology, 2022, 44(1): 149-167.
[13] GAO S H, CHENG M M, ZHAO K, et al. Res2net: a new multi-scale backbone architecture[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2019, 43(2): 652-662.
[14] ROY A M, BOSE R, BHADURI J. A fast accurate fine-grain object detection model based on YOLOv4 deep neural network[J]. Neural Computing and Applications, 2022, 34(5): 3895-3921.
[15] HOU Q, ZHOU D, FENG J. Coordinate attention for efficient mobile network design[C]//Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Nashville, TN, USA, Jun 20-25, 2021. Piscataway: IEEE, 2021: 13713-13722.
[16] HU J, SHEN L, SUN G. Squeeze-and-excitation networks[J]. Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(8): 2011-2023.
[17] WOO S, PARK J, LEE J Y, et al. CBAM: convolutional block attention module[C]//Proceedings of the 15th European Conference on Computer Vision, Munich, Germany, Sep 8-14, 2018. Cham: Springer, 2018: 3-19.
[18] WEN Y, ZHANG K, LI Z, et al. A discriminative feature learning approach for deep face recognition[C]//European Conference on Computer Vision, Amsterdam, The Netherlands, Oct 11-14, 2016. Cham: Springer, 2016: 499-515.
[19] 袁公萍, 汤一平, 韩旺明, 等. 基于深度卷积神经网络的车型识别方法[J]. 浙江大学学报(工学版), 2018, 52(4): 694-702.
YUAN G P, TANG Y P, HAN W M, et al. Vehicle category recognition based on deep convolutional neural network[J]. Journal of Zhejiang University (Engineering Science), 2018, 52(4): 694-702.
[20] HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, Nevada, USA, Jun 27-30, 2016. Piscataway: IEEE, 2016: 770-778.
[21] HUANG G, LIU Z, PLEISS G, et al. Convolutional networks with dense connectivity[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 44(12): 8704-8716.