CoChat
Sign Up
Cited by 90Smart Agriculture and AIAdvanced Neural Network ApplicationsIndustrial Vision Systems and Defect Detection

The DOI matches its Crossref record; 6 of the 6 references it cites were found in a registry or an index.

AgriPest-YOLO: A rapid light-trap agricultural pest detection method based on deep learning

Open access
Zhang, Wei, Huang, He, Sun, Youqiang, Wu, Xiaowei·2022·Frontiers in Plant Science
DOI 10.3389/fpls.2022.1079384PMC PMC9800973
Open this paper in CoChat — free accountOpen PDF

Abstract

Light traps have been widely used for automatic monitoring of pests in the field as an alternative to time-consuming and labor-intensive manual investigations. However, the scale variation, complex background and dense distribution of pests in light-trap images bring challenges to the rapid and accurate detection when utilizing vision technology. To overcome these challenges, in this paper, we put forward a lightweight pest detection model, AgriPest-YOLO, for achieving a well-balanced between efficiency, accuracy and model size for pest detection. Firstly, we propose a coordination and local attention (CLA) mechanism for obtaining richer and smoother pest features as well as reducing the interference of noise, especially for pests with complex backgrounds. Secondly, a novel grouping spatial pyramid pooling fast (GSPPF) is designed, which enriches the multi-scale representation of pest features via fusing multiple receptive fields of different scale features. Finally, soft-NMS is introduced in the prediction layer to optimize the final prediction results of overlapping pests. We evaluated the performance of our method on a large scale multi pest image dataset containing 24 classes and 25k images. Experimental results show that AgriPest-YOLO achieves end-to-end real-time pest detection with high accuracy, obtaining 71.3% mAP on the test dataset, outperforming the classical detection models (Faster RCNN, Cascade RCNN, Dynamic RCNN,YOLOX and YOLOv4) and lightweight detection models (Mobilenetv3-YOLOv4, YOLOv5 and YOLOv4-tiny), meanwhile our method demonstrates better balanced performance in terms of model size, detection speed and accuracy. The method has good accuracy and efficiency in detecting multi-class pests from light-trap images which is a key component of pest forecasting and intelligent pest monitoring technology.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.3389/fpls.2022.1079384

Cites

6 of 6 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Some examples of pest images in Pest24, the red boxes are the pests that need to be predicted; pest detection from light traps is affected by various factors, (A-C, E, F): non-target pest background, (C, F): accidental occlusion, (B, D): oversized non-target background, (D): inflection points caused by light, (E, F): dense distribution.
Figure 1 Some examples of pest images in Pest24, the red boxes are the pests that need to be predicted; pest detection from light traps is affected by various factors, (A-C, E, F): non-target pest background, (C, F): accidental occlusion, (B, D): oversized non-target background, (D): inflection points caused by light, (E, F): dense distribution.CC BY · PubMed Central
Figure 2 Examples of different data augmentation methods; HSV: color-space augmentation; Flip: the image is flipped with some probability; Translate: the image is translated with some proportion; Scale: the image is scaled within some range; Mosaic: the four training images are stitched together after random cropping.
Figure 2 Examples of different data augmentation methods; HSV: color-space augmentation; Flip: the image is flipped with some probability; Translate: the image is translated with some proportion; Scale: the image is scaled within some range; Mosaic: the four training images are stitched together after random cropping.CC BY · PubMed Central
Figure 3 The principle of YOLO detection; X,Y are the center coordinates of the prediction bounding box; H, W are the length and width of the prediction bounding box; Conf, Pred are the confidence score and class probability, respectively.
Figure 3 The principle of YOLO detection; X,Y are the center coordinates of the prediction bounding box; H, W are the length and width of the prediction bounding box; Conf, Pred are the confidence score and class probability, respectively.CC BY · PubMed Central
Figure 4 Pest detection network AgriPest-YOLO structure, consisting mainly of backbone network, GSPPF, neck and head.
Figure 4 Pest detection network AgriPest-YOLO structure, consisting mainly of backbone network, GSPPF, neck and head.CC BY · PubMed Central
Figure 5 Architecture of CLA module; VGP and HGP are vertical and horizontal pooling operations respectively; BN is Batch Normalization; LReLU is the Leaky ReLu activation function.
Figure 5 Architecture of CLA module; VGP and HGP are vertical and horizontal pooling operations respectively; BN is Batch Normalization; LReLU is the Leaky ReLu activation function.CC BY · PubMed Central
Figure 6 Architecture of GSPPF module, MaxPool2D represents the max pooling operation with the kernel size of 5.
Figure 6 Architecture of GSPPF module, MaxPool2D represents the max pooling operation with the kernel size of 5.CC BY · PubMed Central
Figure 7 Detection results of different models for pests with few samples and extremely small pests. (A, B) illustrate the detection results under different IoU thresholds respectively.
Figure 7 Detection results of different models for pests with few samples and extremely small pests. (A, B) illustrate the detection results under different IoU thresholds respectively.CC BY · PubMed Central
Figure 8 Detection results of different models for pests in densely distributed and complex backgrounds. Test_complex is the test set for complex background pests and test_dense is the test set for densely distributed pests. (A, B) illustrate the detection results under different IoU thresholds respectively.
Figure 8 Detection results of different models for pests in densely distributed and complex backgrounds. Test_complex is the test set for complex background pests and test_dense is the test set for densely distributed pests. (A, B) illustrate the detection results under different IoU thresholds respectively.CC BY · PubMed Central
Figure 9 Detection results of YOLOv5s and the proposed AgriPest-YOLO. (A, D, E, H) belongs to the test_complex test set, (B, C, F, G) belongs to the test_dense test set.
Figure 9 Detection results of YOLOv5s and the proposed AgriPest-YOLO. (A, D, E, H) belongs to the test_complex test set, (B, C, F, G) belongs to the test_dense test set.CC BY · PubMed Central
Figure 10 Detection results of YOLOv5s and the proposed AgriPest-YOLO at different noise levels. The horizontal axis indicates the noise intensity added to the test set images and the vertical axis indicates the detection accuracy of the models. (A, B) illustrate the detection results under different IoU thresholds respectively.
Figure 10 Detection results of YOLOv5s and the proposed AgriPest-YOLO at different noise levels. The horizontal axis indicates the noise intensity added to the test set images and the vertical axis indicates the detection accuracy of the models. (A, B) illustrate the detection results under different IoU thresholds respectively.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 6 references

Research works printed but not linked: 23

  1. Cai, 2021, Cascade r-CNN: High quality object detection and instance segmentation, IEEE Trans. Pattern Anal. Mach. Intell., 43, 1483, 10.1109/TPAMI.2019.2956516 · DOI
  2. Gao, 2021, Res2Net: A new multi-scale backbone architecture, IEEE Trans. Pattern Anal. Mach. Intell., 43, 652, 10.1109/TPAMI.2019.2938758 · DOI
  3. Ge, 2021, YOLOX: Exceeding YOLO series in 2021, 10.48550/arXiv.2107.08430 · DOI · Paper page
  4. Glenn, 2020, yolov5
  5. Jiao, 2020, AF-RCNN: An anchor-free convolutional neural network for multi-categories agricultural pest detection, Comput. Electron. Agric., 174, 10.1016/j.compag.2020.105522 · DOI
  6. Jiao, 2022, Global context-Aware-Based deformable residual network module for precise pest recognition and detection, Front. Plant Sci., 13, 10.3389/fpls.2022.895944 · DOI
  7. Jiao, 2022, Adaptive feature fusion pyramid network for multi-classes agricultural pest detection, Comput. Electron. Agric., 195, 10.1016/j.compag.2022.106827 · DOI
  8. Li, 2021, A real-time table grape detection method based on improved YOLOv4-tiny network in complex background, Biosyst. Eng., 212, 347, 10.1016/j.biosystemseng.2021.11.011 · DOI
  9. Li, 2021, Fast and accurate green pepper detection in complex backgrounds via an improved Yolov4-tiny model, Comput. Electron. Agric., 191, 10.1016/j.compag.2021.106503 · DOI
  10. Liu, 2015, ParseNet: Looking wider to see better, 10.48550/arXiv.1506.04579 · DOI
  11. Liu, 2022, A global activated feature pyramid network for tiny pest detection in the wild, Mach. Vision Appl., 33, 76, 10.1007/s00138-022-01310-0 · DOI
  12. Li, 2021, Field detection of tiny pests from sticky trap images using deep learning in agricultural greenhouse, Comput. Electron. Agric., 183, 10.1016/j.compag.2021.106048 · DOI
  13. Li, 2022, A multi-branch convolutional neural network with density map for aphid counting, Biosyst. Eng., 213, 148, 10.1016/j.biosystemseng.2021.11.020 · DOI
  14. Redmon, 2018, YOLOv3: An incremental improvement, 10.48550/arXiv.1804.02767 · DOI · Paper page
  15. Wang, 2022, An efficient attention module for instance segmentation network in pest monitoring, Comput. Electron. Agric., 195, 10.1016/j.compag.2022.106853 · DOI
  16. Wang, 2020, Pest24: A large-scale very small object data set of agricultural pests for multi-target detection, Comput. Electron. Agric., 175, 10.1016/j.compag.2020.105585 · DOI
  17. Wen, 2009, Local feature-based identification and classification for orchard insects, Biosyst. Eng., 104, 299, 10.1016/j.biosystemseng.2009.07.002 · DOI
  18. Xie, 2021, A CNN-based lightweight ensemble model for detecting defective carrots, Biosyst. Eng., 208, 287, 10.1016/j.biosystemseng.2021.06.008 · DOI
  19. Xie, 2015, Automatic classification for field crop insects via multiple-task sparse representation and multiple-kernel learning, Comput. Electron. Agric., 119, 123, 10.1016/j.compag.2015.10.015 · DOI
  20. Yao, 2012, An insect imaging system to automate rice light-trap pest identification, J. Integr. Agric., 11, 978, 10.1016/S2095-3119(12)60089-6 · DOI
  21. Zhang, 2020, Dynamic r-CNN: Towards high quality object detection via dynamic training, 260, 10.1007/978-3-030-58555-6_16 · DOI
  22. Zhang, 2022, Pest region detection in complex backgrounds via contextual information and multi-scale mixed attention mechanism, Agriculture, 12, 10.3390/agriculture12081104 · DOI
  23. Zhao, 2022, Multiple disease detection method for greenhouse-cultivated strawberry based on multiscale feature fusion faster R_CNN, Comput. Electron. Agric., 199, 10.1016/j.compag.2022.107176 · DOI

Cited by

Sorted by citations

See all 90

Metadata

Authors
Zhang, W., Huang, H., Sun, Y., & Wu, X.
Affiliations
  • Anhui University (CN)
  • Chinese Academy of Sciences (CN)
Corresponding author
  • Huang, He
  • Sun, Youqiang
ORCID
Venue
Frontiers in Plant Science
Publisher
Frontiers Media SA
Published
2022-12-16
Volume
13
Pages
1079384-1079384
ISSN
1664-462X (ISSN-L)
DOI
10.3389/fpls.2022.1079384
PMC
PMC9800973
OpenAlex
W4313275642
Cited by
90
References
6
Citations by year
2026: 28 · 2025: 39 · 2024: 14 · 2023: 9
FWCI
8.54
Type
journal article
Language
en
Open access
Licence
Locations
  • Frontiers in Plant Science (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
Chinese Academy of Sciences (XDA28120400)
Topics
  • Life Sciences › Agricultural and Biological Sciences › Plant Science › Smart Agriculture and AI
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
  • Physical Sciences › Engineering › Industrial and Manufacturing Engineering › Industrial Vision Systems and Defect Detection
Keywords
agricultural pest detection, YOLO, lightweight object detection, spatial pyramid pooling, Soft-NMS, multi-scale feature fusion, real-time detection, pest monitoring

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: AgriPest-YOLO: A rapid light-trap agricultural pest detection method based on deep learning. https://app.cochat.ai/papers/10.3389/fpls.2022.1079384
The paper
Zhang, W., Huang, H., Sun, Y., & Wu, X. (2022). AgriPest-YOLO: A rapid light-trap agricultural pest detection method based on deep learning. Frontiers in Plant Science. https://doi.org/10.3389/fpls.2022.1079384