Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Oct 31:15:1398277.
doi: 10.3389/fpls.2024.1398277. eCollection 2024.

LCGSC-YOLO: a lightweight apple leaf diseases detection method based on LCNet and GSConv module under YOLO framework

Jianlong Wang  1 , Congcong Qin  1 , Beibei Hou  1 , Yuan Yuan  2 , Yake Zhang  3 , Wenfeng Feng  1
Affiliations

LCGSC-YOLO: a lightweight apple leaf diseases detection method based on LCNet and GSConv module under YOLO framework

Jianlong Wang et al. Front Plant Sci. .

Abstract

Introduction: In response to the current mainstream deep learning detection methods with a large number of learned parameters and the complexity of apple leaf disease scenarios, the paper proposes a lightweight method and names it LCGSC-YOLO. This method is based on the LCNet(A Lightweight CPU Convolutional Neural Network) and GSConv(Group Shuffle Convolution) module modified YOLO(You Only Look Once) framework.

Methods: Firstly, the lightweight LCNet is utilized to reconstruct the backbone network, with the purpose of reducing the number of parameters and computations of the model. Secondly, the GSConv module and the VOVGSCSP (Slim-neck by GSConv) module are introduced in the neck network, which makes it possible to minimize the number of model parameters and computations while guaranteeing the fusion capability among the different feature layers. Finally, coordinate attention is embedded in the tail of the backbone and after each VOVGSCSP module to improve the problem of detection accuracy degradation issue caused by model lightweighting.

Results: The experimental results show the LCGSC-YOLO can achieve an excellent detection performance with mean average precision of 95.5% and detection speed of 53 frames per second (FPS) on the mixed datasets of Plant Pathology 2021 (FGVC8) and AppleLeaf9.

Discussion: The number of parameters and Floating Point Operations (FLOPs) of the LCGSC-YOLO are much less thanother related comparative experimental algorithms.

Keywords: YOLO; apple leaf disease detection; coordinate attention; depth-wise separable convolution; lightweight network.

PubMed Disclaimer

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

Figure 1
Figure 1
The representative images of different disease categories. (A) Scab. (B) Rust. (C) Powdery_mildew. (D) leaf spot. (E) Altermaria leaf spot. (F) Grey spot. (G) Mosaic. (H) Rust+ Frog_eye_leaf_spot. (I) Scab+ Frog_eye_leaf_spot.
Figure 2
Figure 2
The framework of the two models. (A) The YOLO model. (B) The proposed LCGSC-YOLO model.
Figure 3
Figure 3
The structure of the LCNet module. (A) Depth-wise Convolution. (B) Pointwise Convolution. (C) SE Block.
Figure 4
Figure 4
The structure of GSConv and VOVGSCSP modules. (A) GSConv. (B) The channel shuffle operation. (C) GS bottleneck. (D) VOVGSCSP. (E) Modules relationships.
Figure 5
Figure 5
The structure of coordinate attention.
Figure 6
Figure 6
Radar plots showing the results of the five model tests.
Figure 7
Figure 7
3D bar graphs of test results for five different models.
Figure 8
Figure 8
Comparison of different models for detecting apple leaf disease images. (A-G) indicates the name of the different diseases. The red circles show the contrasting positions.
Figure 9
Figure 9
Comparison of the detection effects of different models on apple leaf disease images in special scenes. The different scenarios are represented from top to bottom.
See this image and copyright information in PMC

References

    1. Ahmed K., Shahidi T. R., Alam S. M. I., Momen S. (2019). “Rice leaf disease detection using machine learning techniques,” in 2019 International Conference on Sustainable Technologies for Industry 4.0 (STI), (Dhaka, Bangladesh: IEEE; ), 01–05.
    1. Arsenovic M., Karanovic M., Sladojevic S., Anderla A., Stefanovic D. (2019). Solving current limitations of deep learning based approaches for plant disease detection. Symmetry 11, 01–18. doi: 10.3390/sym11070939 - DOI
    1. Attri I., Awasthi L. K., Sharma T. P., Rathee P. (2023). A review of deep learning techniques used in agriculture. Ecol. Inf. 77, 01–22. doi: 10.1016/j.ecoinf.2023.102217 - DOI
    1. Barman U., Choudhury R. D., Sahu D., Barman G. G. (2020). Comparison of convolution neural networks for smartphone image based real time classification of citrus leaf disease. Comput. Electron. Agric. 177, 01–09. doi: 10.1016/j.compag.2020.105661 - DOI
    1. Bhuiyan M. A. B., Abdullah H. M., Arman S. E., Rahman S. S., Al Mahmud K. (2023). Bananasqueezenet: A very fast, lightweight convolutional neural network for the diagnosis of three prominent banana leaf diseases. Smart Agric. Technol. 4, 01–13. doi: 10.1016/j.atech.2023.100214 - DOI

LinkOut - more resources