Few-shot object detection for pest insects via features aggregation and contrastive learning
Accurate detection of pest insects is critical for agricultural pest management and crop yield protection, yet traditional detection methods struggle due to the vast diversity of pest species, significant individual differences, and limited labeled data. These challenges are compounded by the typica...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-06-01
|
| Series: | Frontiers in Plant Science |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fpls.2025.1522510/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849688220036497408 |
|---|---|
| author | Shuqian He Shuqian He Biao Jin Biao Jin Xuechao Sun Wenjuan Jiang Wenjuan Jiang Jiaxing Gu Fenglin Gu |
| author_facet | Shuqian He Shuqian He Biao Jin Biao Jin Xuechao Sun Wenjuan Jiang Wenjuan Jiang Jiaxing Gu Fenglin Gu |
| author_sort | Shuqian He |
| collection | DOAJ |
| description | Accurate detection of pest insects is critical for agricultural pest management and crop yield protection, yet traditional detection methods struggle due to the vast diversity of pest species, significant individual differences, and limited labeled data. These challenges are compounded by the typically small size of pest targets and complex environmental conditions. To address these limitations, this study proposes a novel few-shot object detection (FSOD) method leveraging feature aggregation and supervised contrastive learning (SCL) within the Faster R-CNN framework. Our methodology involves multi-scale feature extraction using a Feature Pyramid Network (FPN), enabling the capture of rich semantic information across various scales. A Feature Aggregation Module (FAM) with an attention mechanism is designed to effectively fuse contextual features from support and query images, enhancing representation capabilities for multi-scale and few-sample pest targets. Additionally, supervised contrastive learning is employed to strengthen intra-class similarity and inter-class dissimilarity, thereby improving discriminative power. To manage class imbalance and enhance the focus on challenging samples, focal loss and class weights are integrated into the model’s comprehensive loss function. Experimental validation on the PestDet20 dataset, consisting of diverse tropical pest insects, demonstrates that the proposed method significantly outperforms existing approaches, including YOLO, TFA, VFA, and FSCE. Specifically, our model achieves superior mean Average Precision (mAP) results across different few-shot scenarios (3-shot, 5-shot, and 10-shot), demonstrating robustness and stability. Ablation studies confirm that each component of our method substantially contributes to performance improvement. This research provides a practical and efficient solution for pest detection under challenging conditions, reducing dependency on large annotated datasets and improving detection accuracy for minority pest classes. While computational complexity remains higher than real-time frameworks like YOLO, the significant gains in detection accuracy justify the trade-off for critical pest management applications. |
| format | Article |
| id | doaj-art-2af095cf33c44369a5f566d4c18e01bc |
| institution | DOAJ |
| issn | 1664-462X |
| language | English |
| publishDate | 2025-06-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Plant Science |
| spelling | doaj-art-2af095cf33c44369a5f566d4c18e01bc2025-08-20T03:22:04ZengFrontiers Media S.A.Frontiers in Plant Science1664-462X2025-06-011610.3389/fpls.2025.15225101522510Few-shot object detection for pest insects via features aggregation and contrastive learningShuqian He0Shuqian He1Biao Jin2Biao Jin3Xuechao Sun4Wenjuan Jiang5Wenjuan Jiang6Jiaxing Gu7Fenglin Gu8School of Information Science and Technology, Hainan Normal University, Haikou, Hainan, ChinaHainan Provincial Engineering Research Center for Artificial Intelligence and Equipment for Monitoring Tropical Biodiversity and Ecological Environment, Hainan Normal University, Haikou, Hainan, ChinaSchool of Information Science and Technology, Hainan Normal University, Haikou, Hainan, ChinaHainan Provincial Engineering Research Center for Artificial Intelligence and Equipment for Monitoring Tropical Biodiversity and Ecological Environment, Hainan Normal University, Haikou, Hainan, ChinaHainan Provincial Engineering Research Center for Artificial Intelligence and Equipment for Monitoring Tropical Biodiversity and Ecological Environment, Hainan Normal University, Haikou, Hainan, ChinaSchool of Information Science and Technology, Hainan Normal University, Haikou, Hainan, ChinaHainan Provincial Engineering Research Center for Artificial Intelligence and Equipment for Monitoring Tropical Biodiversity and Ecological Environment, Hainan Normal University, Haikou, Hainan, ChinaCollege of Computer Science and Technology, Zhejiang University, Hangzhou, Zhejiang, ChinaSpice and Beverage Research Institute, Chinese Academy of Tropical Agricultural Sciences, Wanning, Hainan, ChinaAccurate detection of pest insects is critical for agricultural pest management and crop yield protection, yet traditional detection methods struggle due to the vast diversity of pest species, significant individual differences, and limited labeled data. These challenges are compounded by the typically small size of pest targets and complex environmental conditions. To address these limitations, this study proposes a novel few-shot object detection (FSOD) method leveraging feature aggregation and supervised contrastive learning (SCL) within the Faster R-CNN framework. Our methodology involves multi-scale feature extraction using a Feature Pyramid Network (FPN), enabling the capture of rich semantic information across various scales. A Feature Aggregation Module (FAM) with an attention mechanism is designed to effectively fuse contextual features from support and query images, enhancing representation capabilities for multi-scale and few-sample pest targets. Additionally, supervised contrastive learning is employed to strengthen intra-class similarity and inter-class dissimilarity, thereby improving discriminative power. To manage class imbalance and enhance the focus on challenging samples, focal loss and class weights are integrated into the model’s comprehensive loss function. Experimental validation on the PestDet20 dataset, consisting of diverse tropical pest insects, demonstrates that the proposed method significantly outperforms existing approaches, including YOLO, TFA, VFA, and FSCE. Specifically, our model achieves superior mean Average Precision (mAP) results across different few-shot scenarios (3-shot, 5-shot, and 10-shot), demonstrating robustness and stability. Ablation studies confirm that each component of our method substantially contributes to performance improvement. This research provides a practical and efficient solution for pest detection under challenging conditions, reducing dependency on large annotated datasets and improving detection accuracy for minority pest classes. While computational complexity remains higher than real-time frameworks like YOLO, the significant gains in detection accuracy justify the trade-off for critical pest management applications.https://www.frontiersin.org/articles/10.3389/fpls.2025.1522510/fullfeature aggregationcontrastive learningfew-shot learningobject detectionpest control |
| spellingShingle | Shuqian He Shuqian He Biao Jin Biao Jin Xuechao Sun Wenjuan Jiang Wenjuan Jiang Jiaxing Gu Fenglin Gu Few-shot object detection for pest insects via features aggregation and contrastive learning Frontiers in Plant Science feature aggregation contrastive learning few-shot learning object detection pest control |
| title | Few-shot object detection for pest insects via features aggregation and contrastive learning |
| title_full | Few-shot object detection for pest insects via features aggregation and contrastive learning |
| title_fullStr | Few-shot object detection for pest insects via features aggregation and contrastive learning |
| title_full_unstemmed | Few-shot object detection for pest insects via features aggregation and contrastive learning |
| title_short | Few-shot object detection for pest insects via features aggregation and contrastive learning |
| title_sort | few shot object detection for pest insects via features aggregation and contrastive learning |
| topic | feature aggregation contrastive learning few-shot learning object detection pest control |
| url | https://www.frontiersin.org/articles/10.3389/fpls.2025.1522510/full |
| work_keys_str_mv | AT shuqianhe fewshotobjectdetectionforpestinsectsviafeaturesaggregationandcontrastivelearning AT shuqianhe fewshotobjectdetectionforpestinsectsviafeaturesaggregationandcontrastivelearning AT biaojin fewshotobjectdetectionforpestinsectsviafeaturesaggregationandcontrastivelearning AT biaojin fewshotobjectdetectionforpestinsectsviafeaturesaggregationandcontrastivelearning AT xuechaosun fewshotobjectdetectionforpestinsectsviafeaturesaggregationandcontrastivelearning AT wenjuanjiang fewshotobjectdetectionforpestinsectsviafeaturesaggregationandcontrastivelearning AT wenjuanjiang fewshotobjectdetectionforpestinsectsviafeaturesaggregationandcontrastivelearning AT jiaxinggu fewshotobjectdetectionforpestinsectsviafeaturesaggregationandcontrastivelearning AT fenglingu fewshotobjectdetectionforpestinsectsviafeaturesaggregationandcontrastivelearning |