Thrips detection on sticky traps based on improved YOLOv11n
4 affiliations
- 1State Key Laboratory for Quality and Safety of Agro-products, Xianghu Laboratory, Institute of Plant Protection and Microbiology, Zhejiang Academy of Agricultural Sciences
- 2Key Laboratory of Tea Quality and Safety Control, Ministry of Agriculture and Rural Affairs, Tea Research Institute, Chinese Academy of Agricultural Sciences
- 3Hangzhou Ruikun Technology Co., Ltd.
- 4Anji County Agriculture and Rural Affairs Bureau
Abstract
Thrips on sticky traps are extremely small and densely distributed, making automated detection challenging. This study proposes YOLOv11n-WNE, an improved YOLOv11n-based detector tailored for thrips detection in greenhouse environments. Three modifications are introduced: (1) a Wavelet Unpooling StarFusion (WSF) network in the neck, combining WaveletUnPool upsampling—which uses the discrete wavelet transform to preserve spatial details—with StarFusion feature enhancement that captures high-order multi-scale interactions through element-wise multiplication; (2) a Normalized Wasserstein Distance (NWD) loss that models bounding boxes as 2-D Gaussian distributions, providing smooth gradient responses for tiny-object localization; and (3) an EfficientHead detection head using grouped convolutions (G = 8) to compress parameters and computation. A dataset of 2, 400 images with 40, 480 annotated thrips instances was constructed for training (1, 680), validation (360), and testing (360). YOLOv11n-WNE achieves [email protected] of 91.5%, with precision of 88.2% and recall of 85.0%—improvements of 4.1, 4.5, and 5.1 percentage points over the baseline YOLOv11n, respectively—while reducing parameters to 2.11M (81.8% of baseline) and model size to 4.5 MB. On the GPU, the model runs at 123.8 FPS at an input size of 640×640, supporting its deployment for edge-based intelligent thrips monitoring in greenhouses. Beyond individual novelty, the contribution lies in problem-driven integration. WSF, NWD, and EfficientHead act on complementary stages of the detection pipeline—feature aggregation, localization supervision, and model compression, respectively. Together they address the failure modes of thrips-on-sticky-trap imagery: dense tiny objects, weak textures, and frequent overlap with non-target insects.
Related
- Complex scene green walnut target detection method based on improved YOLOv11n
- Advantages and Limitations of Yellow Sticky Traps as a Pest Control Method in Pollinator-Friendly Rapeseed Agroecosystems: A Pilot Study in Northern Italy
- Detection of Neutrophil Extracellular Traps by Flow Cytometry in Healthy Controls and Patients with Systemic Lupus Erythematosus
- Comparative Analysis of Thrips Prey and Pollen Usage by Three Predatory Mites, Amblyseius largoensis, Neoseiulus californicus and Neoseiulus barkeri
- Improved detection and spatiotemporal spectral analysis of neural traveling waves
BibTeX
@article{zhou2026thrips,
title = {{Thrips detection on sticky traps based on improved YOLOv11n}},
author = {Shuxing Zhou and Yang Li and Yuheng Lu and Haoyang Liu and Jun Li and Yuhan Pan and Xiaowei Li and Jinming Zhang and Zhijun Zhang and Jun Huang and Wanying Dong and Yaobin Lu},
journal = {Frontiers in Plant Science},
year = {2026},
month = oct,
doi = {10.3389/fpls.2026.1945602},
url = {https://www.frontiersin.org/articles/10.3389/fpls.2026.1945602/full},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Thrips detection on sticky traps based on improved YOLOv11n
AU - Shuxing Zhou
AU - Yang Li
AU - Yuheng Lu
AU - Haoyang Liu
AU - Jun Li
AU - Yuhan Pan
AU - Xiaowei Li
AU - Jinming Zhang
AU - Zhijun Zhang
AU - Jun Huang
AU - Wanying Dong
AU - Yaobin Lu
PY - 2026
DA - 2026/10/29
JO - Frontiers in Plant Science
DO - 10.3389/fpls.2026.1945602
UR - https://www.frontiersin.org/articles/10.3389/fpls.2026.1945602/full
AB - Thrips on sticky traps are extremely small and densely distributed, making automated detection challenging. This study proposes YOLOv11n-WNE, an improved YOLOv11n-based detector tailored for thrips detection in greenhouse environments. Three modifications are introduced: (1) a Wavelet Unpooling StarFusion (WSF) network in the neck, combining WaveletUnPool upsampling—which uses the discrete wavelet transform to preserve spatial details—with StarFusion feature enhancement that captures high-order multi-scale interactions through element-wise multiplication; (2) a Normalized Wasserstein Distance (NWD) loss that models bounding boxes as 2-D Gaussian distributions, providing smooth gradient responses for tiny-object localization; and (3) an EfficientHead detection head using grouped convolutions (G = 8) to compress parameters and computation. A dataset of 2, 400 images with 40, 480 annotated thrips instances was constructed for training (1, 680), validation (360), and testing (360). YOLOv11n-WNE achieves [email protected] of 91.5%, with precision of 88.2% and recall of 85.0%—improvements of 4.1, 4.5, and 5.1 percentage points over the baseline YOLOv11n, respectively—while reducing parameters to 2.11M (81.8% of baseline) and model size to 4.5 MB. On the GPU, the model runs at 123.8 FPS at an input size of 640×640, supporting its deployment for edge-based intelligent thrips monitoring in greenhouses. Beyond individual novelty, the contribution lies in problem-driven integration. WSF, NWD, and EfficientHead act on complementary stages of the detection pipeline—feature aggregation, localization supervision, and model compression, respectively. Together they address the failure modes of thrips-on-sticky-trap imagery: dense tiny objects, weak textures, and frequent overlap with non-target insects.
SN - 1664-462X
ER -