Lightweight Semantic Segmentation of Tobacco Main Veins Fusing Coordinate Attention and Dense Connectivity

doi:10.3778/j.issn.1002-8331.2308-0228

Abstract

Abstract: Aiming at the current problem of low automation in the process of analysing the main veins of tobacco leaves, which makes it difficult to cope with the extraction and recognition of complex main veins of tobacco leaves, a lightweight semantic segmentation of the main veins of tobacco leaves by integrating coordinate attention and mixed-connections atrous spatial pyramid pooling (MASPP) is proposed. The algorithm takes DeepLabV3+ network model as the framework, and adopts lightweight MobileNetV2 to replace the Xception network in the original framework, and carries out the main feature extraction in the way of “expanding-extracting-compressing”, so as to reduce the number of parameters of the network model, and introduces the coordinate attention mechanism to strengthen the learning ability of subtle features of main veins, and improves the learning ability of subtle features of main veins, and improves the learning ability of subtle features of main veins of the leaf. The introduction of the coordinate attention mechanism enhances the learning ability of the subtle features of the main veins of the tobacco leaves, and improves the regional misclassification of the main vein segmentation compared with the real distribution of the main veins. The MASPP structure of “mixed-connected dense sampling” is used to replace the empty space convolution pooling pyramid in the original network model, and improves the intermittent segmentation of the main veins of the tobacco leaves. The experimental results show that compared with the original DeepLabV3+ semantic segmentation algorithm, the training time is reduced from 635 min to 311 min, the average interaction ratio (mIOU) reaches 80.66%, the average pixel accuracy (mPA) reaches 91.96%, the number of parameters in the network model is compressed by 85.32%, and the storage space is reduced to 30.63 MB. The network model parameters are compressed by 85.32%, and the storage space is reduced to 30.63 MB.

Key words: main vein of tobacco leaf, light weight, attention mechanism, pyramid pooling, DeepLabV3+

摘要： 针对目前烟叶主脉的分析过程自动化低，难以应对复杂烟叶主脉提取与识别的问题，提出一种基于坐标注意力（coordinate attention，CA）与混合联接空洞空间金字塔池化（mixed-connections atrous spatial pyramid pooling，MASPP）的烟叶主脉轻量级语义分割方法。该算法以DeepLabV3+网络模型为框架，采用轻量级MobileNetV2替换原始框架中的Xception网络，以“扩充-提取-压缩”方式进行主干特征提取，减少网络模型参数量；引入坐标注意力机制加强对烟叶主脉细微特征的学习能力，改善分割主脉时与主脉真实分布相比较所存在的区域错分情况；采用“混联密接采样”的MASPP结构替代原始网络模型中的空洞空间卷积池化金字塔，改善烟叶主脉分割存在的断续分割情况。实验结果表明，与原始的DeepLabV3+语义分割算法相比较，训练时间从635?min缩减为311?min，平均交互比（mIOU）达到80.66%，平均像素精度（mPA）达到91.96%，网络模型参数量压缩85.32%，储存空间降为30.63?MB。在保证分割精度的同时减少模型训练时间，为烟叶主脉分割提供了新的思路和方法。

关键词: 烟叶主脉, 轻量化, 注意力机制, 密接采样, DeepLabV3+

SU Shuailin, GAN Bomin, LONG Jie, LIU Yuchen, GAI Xiaolei, ZHANG Jiwu. Lightweight Semantic Segmentation of Tobacco Main Veins Fusing Coordinate Attention and Dense Connectivity[J]. Computer Engineering and Applications, 2024, 60(24): 250-259.

苏帅林, 甘博敏, 龙杰, 刘宇晨, 盖小雷, 张冀武. 融合坐标注意力与混联采样的烟叶主脉轻量级语义分割[J]. 计算机工程与应用, 2024, 60(24): 250-259.

References

[1] LU M, ZHOU Q, CHEN T, et al. Qualitative discrimination of intact tobacco leaves based on near-infrared technology[J]. Journal of Spectroscopy, 2021: 1-9.
[2] MARZAN C S, MARCOS N. Towards tobacco leaf detection using Haar cascade classifier and image processing techniques[C]//Proceedings of the 2nd International Conference on Graphics and Signal Processing, 2018: 63-68.
[3] SELDA J D S, ELLERA R M R, CAJAYON L C, et al. Plant identification by image processing of leaf veins[C]//Proceedings of the International Conference on Imaging, Signal Processing and Communication, 2017: 40-44.
[4] LINSANGAN N B, PANGANTIHON JR R S. FPGA-based plant identification through leaf veins[C]//Proceedings of the 2018 5th International Conference on Biomedical and Bioinformatics Engineering, 2018: 100-104.
[5] FENG S. Kernel pooling feature representation of pre-trained convolutional neural networks for leaf recognition[J]. Multimedia Tools and Applications, 2022, 81(3): 4255-4282.
[6] CUI S, CHEN H, LIANG F, et al. Veins feature extraction for LED plant growth cabinet[C]//Proceedings of the 2016 35th Chinese Control Conference (CCC), 2016: 4917-4920.
[7] 林挺, 余孝源, 李丰果. 复叶中小叶主叶脉及其生长角的提取[J]. 安徽农业科学, 2014, 42(7): 2178-2180.
LIN T, YU X Y, LI F G. Extraction of leaflet main veins and growth angles of leaflets in compound leaves[J]. Journal of Anhui Agricultural Sciences, 2014, 42(7): 2178-2180.
[8] TIAN T, LIU Q, YIN S W, et al. The vein extraction algorithm based on total variation denoising and eight-direction sobel operator[J]. Acta Agriculturae Zhejiangensis, 2015, 27(4): 678-683.
[9] 翁海勇, 李效彬, 肖康松, 等. 基于Mask R-CNN的柑橘主叶脉显微图像实例分割模型[J]. 农业机械学报, 2023, 54(7): 252-258.
WENG H Y, LI X B, XIAO K S, et al. Instance segmentation model for microscopic image of citrus main leaf vein based on Mask R-CNN[J]. Journal of Agricultural Machinery, 2023, 54(7): 252-258.
[10] FUENTES-PACHECO J, TORRES-OLIVARES J, ROMAN-RANGEL E, et al. Fig plant segmentation from aerial images using a deep convolutional encoder-decoder network[J]. Remote Sensing, 2019, 11(10): 1157.
[11] ZHU H Y, HUANG X Y, ZHANG S P, et al. Plant identification via multipath sparse coding[J]. Multimedia Tools and Applications, 2017, 76(3): 4599-4615.
[12] LIU X X, XU F, SUN Y, et al. Convolutional recurrent neural networks for observation-centered plant identification[J]. Journal of Electrical and Computer Engineering, 2018(1): 9373210.
[13] CHEN Z, TING D, NEWBURY R, et al. Semantic segmentation for partially occluded apple trees based on deep learning[J]. Computers and Electronics in Agriculture, 2021, 181: 105952.
[14] SANDLER M, HOWARD A, ZHU M, et al. MobileNetv2: inverted residuals and linear bottlenecks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 4510-4520.
[15] HOU Q, ZHOU D, FENG J. Coordinate attention for efficient mobile network design[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021: 13713-13722.
[16] KWON S. Att-Net: enhanced emotion recognition system using lightweight self-attention module[J]. Applied Soft Computing, 2021, 102: 107101.
[17] GUO M H, LIU Z N, MU T J, et al. Beyond self-attention: external attention using two linear layers for visual tasks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 45(5): 5436-5447.
[18] LUO H, ZHANG S, LEI M, et al. Simplified self-attention for transformer-based end-to-end speech recognition[C]//Proceedings of the 2021 IEEE Spoken Language Technology Workshop (SLT), 2021: 75-81.
[19] GOU F, WU J. Message transmission strategy based on recurrent neural network and attention mechanism in IoT system[J]. Journal of Circuits, Systems and Computers, 2022, 31(7): 2250126.
[20] HOWARD A, SANDLER M, CHU G, et al. Searching for MobileNetv3[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019: 1314-1324.
[21] ZHAO H, SHI J, QI X, et al. Pyramid scene parsing network[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017: 2881-2890.
[22] RONNEBERGER O, FISCHER P, BROX T. U-net: convolutional networks for biomedical image segmentation[C]//Proceedings of the 18th International Conference on Medical Image Computing and Computer-Assisted Intervention(MICCAI 2015), Munich, Germany, October 5-9, 2015: 234-241.
[23] DIAKOGIANNIS F I, WALDNER F, CACCETTA P, et al. ResUNet-a: a deep learning framework for semantic segmentation of remotely sensed data[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2020, 162: 9.