CBAM-U-Net:基于 U-Net 和自注意力机制的马铃薯花粉图像分割

doi:10.6040/j.issn.1672-3961.0.2025.114

摘要/Abstract

摘要： 为解决马铃薯花粉显微图像分割与计数中传统人工观察和阈值分割方法效率低、主观性强及复杂背景适应性不足等问题,提出一种融合卷积块注意力模块(convolutional block attention module, CBAM)和U-Net的花粉分割方法CBAM-U-Net。以U-Net为基础分割框架,引入CBAM,增强网络对花粉区域关键特征及边缘细节的提取能力,采用中值滤波与直方图均衡化对显微图像进行预处理。在44张高分辨率马铃薯花粉图像和5 000余张显微图像上的试验表明,所提方法能够精确分割花粉区域,分割效率较传统方法提升约5倍,在大规模数据集上具有良好的鲁棒性和泛化能力。所提方法可有效提升马铃薯花粉显微图像分析自动化与精准化水平,为作物育种、细胞生物学及精准农业等领域的显微图像分析提供技术支撑。

关键词: U-Net, CBAM, 马铃薯花粉, 高通量分割, 阈值分割

Abstract: To address the low efficiency, subjectivity, and poor adaptability to complex backgrounds of traditional manual observation and threshold-based methods in potato pollen microscopy image segmentation and counting, a pollen segmentation method integrating the convolutional block attention module(CBAM)and U-Net(CBAM-U-Net)was proposed. Built on the U-Net framework, the CBAM was introduced to improve the extraction of key features and edge details in pollen regions, while median filtering and histogram equalization were applied for image preprocessing. Experiments on 44 high-resolution potato pollen images and more than 5 000 microscopic images showed that the proposed method achieved accurate pollen segmentation, improved segmentation efficiency by about five times over traditional methods, and demonstrated good robustness and generalization ability on large-scale datasets. The proposed method could improve the automation and accuracy of potato pollen microscopy image analysis and provide technical support for microscopic image analysis in crop breeding, cell biology, and precision agriculture.

Key words: U-Net, CBAM, potato pollen, high throughput segmentation, threshold segmentation

中图分类号:

TP391.9

沈亚婕,夏璐,李杰,唐明靖. CBAM-U-Net:基于 U-Net 和自注意力机制的马铃薯花粉图像分割[J]. 山东大学学报 (工学版), 2026, 56(4): 84-93.

Shen Yajie, Xia Lu, Li Jie, Tang Mingjing. CBAM-U-Net: segmentation of potato pollen images based on U-Net and self-attention mechanism[J]. Journal of Shandong University(Engineering Science), 2026, 56(4): 84-93.

参考文献

[1] Singh A, Ganapathysubramanian B, Singh A K, et al. Machine learning for high-throughput stress phenotyping in plants[J]. Trends in Plant Science, 2016, 21(2): 110-124.
[2] Lin E, Lane H Y. Machine learning and systems genomics approaches for multi-omics data[J]. Biomarker Research, 2017, 5(1): 2.
[3] Tester M, Langridge P. Breeding technologies to increase crop production in a changing world[J]. Science, 2010, 327(5967): 818-822.
[4] Ribaut J M, De Vicente M C, Delannay X. Molecular breeding in developing countries: challenges and perspectives[J]. Current Opinion in Plant Biology, 2010, 13(2): 213-218.
[5] Boavida L C, Vieira A M, Becker J D, et al. Gametophyte interaction and sexual reproduction: how plants make a zygote[J]. The International Journal of Developmental Biology, 2005, 49(5/6): 615-632.
[6] Fetter K C, Eberhardt S, Barclay R S, et al. StomataCounter: a neural network for automatic stomata identification and counting[J]. New Phytologist, 2019, 223(3): 1671-1681.
[7] Krizhevsky A, Sutskever I, Hinton G E. ImageNet classification with deep convolutional neural networks[J]. Communications of the ACM, 2017, 60(6): 84-90.
[8] Ronneberger O, Fischer P, Brox T. U-Net: convolutional networks for biomedical image segmentation[C] //Medical Image Computing and Computer-Assisted Intervention: MICCAI 2015. Cham, Switzerland: Springer, 2015: 234-241.
[9] Colmer J, O'neill C M, Wells R, et al. SeedGerm: a cost-effective phenotyping platform for automated seed imaging and machine-learning based phenotypic analysis of crop seed germination[J]. New Phytologist, 2020, 228(2): 778-793.
[10] Tello J, Montemayor M I, Forneck A, et al. A new image-based tool for the high throughput phenotyping of pollen viability: evaluation of inter- and intra-cultivar diversity in grapevine[J]. Plant Methods, 2018, 14: 3.
[11] Maraci M A, Bridge C P, Napolitano R, et al. A framework for analysis of linear ultrasound videos to detect fetal presentation and heartbeat[J]. Medical Image Analysis, 2017, 37: 22-36.
[12] Zhao Z X, Chen K X, Yamane S. CBAM-Unet++: easier to find the target with the attention module "CBAM"[C] //2021 IEEE 10th Global Conference on Consumer Electronics(GCCE). Kyoto, Japan: IEEE, 2021: 655-657.
[13] Taud H, Mas J F. Multilayer perceptron(MLP)[M] // Camacho Olmedo M T, Paegelow M, Mas J F, et al. Geomatic approaches for modeling land change scenarios. Cham, Switzerland: Springer, 2017: 451-455.
[14] 周逸凡, 张灵维, 周正东, 等. 基于注意力机制和深度学习的群体语言想象脑电信号分类[J]. 浙江大学学报(工学版), 2024, 58(12): 2540-2546. Zhou Yifan, Zhang Lingwei, Zhou Zhengdong, et al. Classification of group speech imagined EEG signals based on attention mechanism and deep learning[J]. Journal of Zhejiang University(Engineering Science), 2024, 58(12): 2540-2546.
[15] 于贺婷, 刘思萌, 文峰. 基于CBAM注意力机制的智能交通信号控制[J]. 沈阳理工大学学报, 2024, 43(5): 34-40. Yu Heting, Liu Simeng, Wen Feng. Intelligent traffic control technology based on CBAM attention mechanism[J]. Journal of Shenyang Ligong University, 2024, 43(5): 34-40.
[16] 冯庆贺. 面向图像检索的底层视觉与深度卷积特征提取方法研究[D]. 沈阳: 东北大学, 2020: 13-15. Feng Qinghe. Research on low-level vision and deep convolutional feature extraction methods for image retrieval[D]. Shenyang: Northeastern University, 2020: 13-15.
[17] 王军, 张霁云, 程勇. 基于边缘特征和注意力机制的图像语义分割[J]. 计算机系统应用, 2024, 33(7): 63-73. Wang Jun, Zhang Jiyun, Cheng Yong. Image semantic segmentation based on edge features and attention mechanism[J]. Computer Systems & Applications, 2024, 33(7): 63-73.
[18] He K M, Zhang X Y, Ren S Q, et al. Deep residual learning for image recognition[C] //2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR). Las Vegas, USA: IEEE, 2016: 770-778.
[19] He K M, Zhang X Y, Ren S Q, et al. Identity mappings in deep residual networks[C] //Computer Vision-ECCV 2016. Cham, Switzerland: Springer, 2016: 630-645.
[20] 李顺勇, 胥瑞, 李师毅. 加入跳跃连接的深度嵌入K-means聚类[J]. 计算机系统应用, 2024, 33(1): 11-21. Li Shunyong, Xu Rui, Li Shiyi. Deep embedded K-means clustering with skip connections[J]. Computer Systems & Applications, 2024, 33(1): 11-21.
[21] 皮磊, 朱磊, 郑翔, 等. 基于改进Wave-U-Net跳跃连接的盲源分离算法[J]. 信号处理, 2022, 38(4): 835-843. Pi Lei, Zhu Lei, Zheng Xiang, et al. Blind source separation algorithm based on improved Wave-U-Net skip connection[J]. Journal of Signal Processing, 2022, 38(4): 835-843.
[22] Creswell A, Bharath A A. Denoising adversarial autoencoders[J]. IEEE Transactions on Neural Networks and Learning Systems, 2019, 30(4): 968-984.
[23] 史加荣, 王丹, 尚凡华, 等. 随机梯度下降算法研究进展[J]. 自动化学报, 2021, 47(9): 2103-2119. Shi Jiarong, Wang Dan, Shang Fanhua, et al. Research advances on stochastic gradient descent algorithms[J]. Acta Automatica Sinica, 2021, 47(9): 2103-2119.
[24] 赵高长, 张磊, 武风波. 改进的中值滤波算法在图像去噪中的应用[J]. 应用光学, 2011, 32(4): 678-682. Zhao Gaochang, Zhang Lei, Wu Fengbo. Application of improved median filtering algorithm to image denoising[J]. Journal of Applied Optics, 2011, 32(4): 678-682.
[25] 毛本清, 金小梅. 自适应直方图均衡化算法在图像增强处理的应用[J]. 河北北方学院学报(自然科学版), 2010, 26(5): 64-68. Mao Benqing, Jin Xiaomei. Application of self-adaptive histogram equalization algorithm to image enhancement processing[J]. Journal of Hebei North University(Natural Science Edition), 2010, 26(5): 64-68.
[26] Van Valen D A, Kudo T, Lane K M, et al. Deep learning automates the quantitative analysis of individual cells in live-cell imaging experiments[J]. PLoS Computational Biology, 2016, 12(11): e1005177.
[27] Liu X M, Zhao D B, Xiong R Q, et al. Image interpolation via regularized local linear regression[J]. IEEE Transactions on Image Processing, 2011, 20(12): 3455-3469.
[28] 彭程, 李帅, 苗艳龙, 等. 基于三维点云的番茄植株茎叶分割与表型特征提取[J]. 农业工程学报, 2022, 38(9): 187-194. Peng Cheng, Li Shuai, Miao Yanlong, et al. Stem-leaf segmentation and phenotypic trait extraction of tomatoes using three-dimensional point cloud[J]. Transactions of the Chinese Society of Agricultural Engineering, 2022, 38(9): 187-194.
[29] Ranefall P, Wählby C. Global gray-level thresholding based on object size[J]. Cytometry Part A, 2016, 89(4): 385-390.
[30] Song J, Jiao W, Lankowicz K, et al. A two-stage adaptive thresholding segmentation for noisy low-contrast images[J]. Ecological Informatics, 2022, 69: 101632.
[31] Ye J, Xu G. Geometric flow approach for region-based image segmentation[J]. IEEE Transactions on Image Processing, 2012, 21(12): 4735-4745.
[32] Phornphatcharaphong W, Eua-Anant N. Edge-based color image segmentation using particle motion in a vector image field derived from local color distance images[J]. Journal of Imaging, 2020, 6(7): 72.
[33] Xing J W, Yang P, Qing G L T. Robust 2D Otsu's algorithm for uneven illumination image seg-mentation[J]. Computational Intelligence and Neuro-science, 2020, 2020(1): 5047976.
[34] Huang S Y, Hsu W L, Hsu R J, et al. Fully convolutional network for the semantic segmentation of medical images: a survey[J]. Diagnostics, 2022, 12(11): 2765.
[35] Badrinarayanan V, Kendall A, Cipolla R. SegNet: a deep convolutional encoder-decoder architecture for image segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(12): 2481-2495.
[36] Stringer C, Wang T, Michaelos M, et al. Cellpose: a generalist algorithm for cellular segmentation[J]. Nature Methods, 2021, 18(1): 100-106.
[37] Stringer C, Pachitariu M. Cellpose3: one-click image restoration for improved cellular segmentation[J]. Nature Methods, 2025, 22(3): 592-599.
[38] 田萱, 王亮, 丁琪. 基于深度学习的图像语义分割方法综述[J]. 软件学报, 2019, 30(2): 440-468. Tian Xuan, Wang Liang, Ding Qi. Review of image semantic segmentation based on deep learning[J]. Journal of Software, 2019, 30(2): 440-468.
[39] 李刚森. 基于深度学习的细胞核图像分割方法研究[D]. 黑龙江: 哈尔滨工业大学, 2018: 13-15. Li Gangsen. Methodology research of nucleus image segmentation based on deep learning[D]. Heilongjiang: Harbin Institute of Technology, 2018: 13-15.
[40] Hu J, Shen L, Sun G. Squeeze-and-excitation networks[C] //2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, USA: IEEE, 2018: 7132-7141.
[41] Shan T, Yan J Y. SCA-Net: a spatial and channel attention network for medical image segmentation[J]. IEEE Access, 2021, 9: 160926-160937.
[42] Hou Q B, Zhou D Q, Feng J S. Coordinate attention for efficient mobile network design[C] //2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR). Nashville, USA: IEEE, 2021: 13708-13717.
[43] Misra D, Nalamada T, Arasanipalai A U, et al. Rotate to attend: convolutional triplet attention module[C] //2021 IEEE Winter Conference on Applications of Computer Vision(WACV). Waikoloa, USA: IEEE, 2021: 3138-3147.
[44] Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need[C] //Proceedings of the 31st Conference on Neural Information Processing Systems(NIPS 2017). Long Beach, USA: Curran Associates, Inc., 2017: 5998-6008.
[45] 赵海丽, 包大泱, 张从豪, 等. 基于改进SDU-YOLOv8的军事飞机目标检测算法[J]. 兵工学报, 2026, 47(1): 250294. Zhao Haili, Bao Dayang, Zhang Conghao, et al. Military aircraft object detection algorithm based on improved SDU-YOLOv8[J]. Acta Armamentarii, 2026, 47(1): 250294.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed