Search

Research on Plant Seedling Image Recognition Based on CBAM-EfficientNet-B0

Author(s)

Ke Zhang1, Ruoxi Cui2, Xiaofeng Li3,*

Affiliation(s)

1College of Artificial Intelligence and Big Data, Henan University of Technology, Zhengzhou, Henan, China 2 College of Information Science and Engineering, Henan University of Technology, Zhengzhou, Henan, China 3ANHUI USTC iFLYTEK Co., Ltd, Hefei, Anhui, China *Corresponding Author

Abstract

With the ongoing development of smart agriculture and computer vision, deep learning based approaches for plant recognition have gradually emerged as an important research direction in agricultural intelligence. During the seedling stage, different plant categories often exhibit similar morphological traits and background interference is relatively high. Moreover, traditional convolutional neural networks lack sufficient feature extraction ability. Therefore, we propose a plant seedling image recognition method based on CBA MEfficientNetB0. Using EfficientNetB0 as the backbone network, we integrate a Convolutional Block Attention Module after each MBConv block. By leveraging both channel and spatial attention mechanisms, the proposed model can more effectively concentrate on critical feature regions, thereby enhancing classification performance for plant seedling images. Our experimental results demonstrate that the combination of attention mechanisms with a lightweight convolutional neural network can effectively improve plant seedling image recognition, and also has practical utility for automatic plant identification and classification in smart agriculture.

Keywords

Plant Seedling Recognition; Deep Learning; EfficientNet-B0; CBAM Attention Mechanism; Image Classification

References

[1] DANG W Y, WANG Z L. Practical application of computer and IoT technology in smart agriculture. Southern Agricultural Machinery, 2026, 57(7): 170-173. [2] ZHOU W Z, YANG T R, ZHONG Z H. Plant phenotypic feature image recognition system based on deep learning. Computer Knowledge and Technology, 2024, 20(36): 49-52. DOI: 10.14004/j.cnki.ckt.2024.1863. [3] ZHOU Z H. Machine Learning. Beijing: Tsinghua University Press, 2016. [4] QIU X P. Neural Networks and Deep Learning. Beijing: Publishing House of Electronics Industry, 2020. [5] Tan M, Le V Q. EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. CoRR, 2019, abs/1905.11946 [6] Shin M, Seo J, Lee B I, et al. Defective Photovoltaic Module Detection Using EfficientNet-B0 in the MachineVision Environment. Machines, 2026, 14 (2): 232-232. DOI: 10.3390/MACHINES14020232. [7] Juseong L, Hoyoung T, Jongsun P. Energy Efficient Canny Edge Detector for Advanced Mobile Vision Applications. IEEE Transactions on Circuits and Systems for Video Technology, 2018, 28 (4): 1037-1046. DOI:10.1109/tcsvt.2016.2640038. [8] Jiang P, Zhang J, Chen J. Enhanced rain removal network with convolutional block attention module (CBAM): a novel approach to image de-raining. EURASIP Journal on Advances in Signal Processing, 2025, 2025 (1): 9-9. DOI: 10.1186/S13634-025-01212-Z. [9] YANG H K, ZHU B W, ZHANG Y M, et al. A review of plant disease image recognition algorithms based on deep learning. Application of Electronic Technique, 2025, 51(1): 1-7. DOI: 10.16157/j.issn.0258-7998.245285. [10]WEI J A. Application of computer vision technology in image data processing. Popular Standardization, 2026, (5): 149-151. [11]Yao J, Liu H, Ye J. Enhanced EfficientNet-B0 with Dual Attention Mechanisms for Food Category Classification in X-ray Images. Journal of Nondestructive Evaluation, 2026, 45 (2): 55-55. DOI: 10.1007/S10921-026-01348-4.