GC-Faster RCNN: The Object Detection Algorithm for Agricultural Pests Based on Improved Hybrid Attention Mechanism
Pest infestations remain a critical threat to global agriculture, significantly compromising crop yield and quality. While accurate pest detection forms the foundation of precision pest management, current approaches face two primary challenges: (1) the scarcity of comprehensive multi-scale, multi-c...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-04-01
|
| Series: | Plants |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2223-7747/14/7/1106 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850213173415641088 |
|---|---|
| author | Bolun Guan Yaqian Wu Jingbo Zhu Juanjuan Kong Wei Dong |
| author_facet | Bolun Guan Yaqian Wu Jingbo Zhu Juanjuan Kong Wei Dong |
| author_sort | Bolun Guan |
| collection | DOAJ |
| description | Pest infestations remain a critical threat to global agriculture, significantly compromising crop yield and quality. While accurate pest detection forms the foundation of precision pest management, current approaches face two primary challenges: (1) the scarcity of comprehensive multi-scale, multi-category pest datasets and (2) performance limitations in detection models caused by substantial target scale variations and high inter-class morphological similarity. To address these issues, we present three key contributions: First, we introduce Insect25—a novel agricultural pest detection dataset containing 25 distinct pest categories, comprising 18,349 high-resolution images. This dataset specifically addresses scale diversity through multi-resolution acquisition protocols, significantly enriching feature distribution for robust model training. Second, we propose GC-Faster RCNN, an enhanced detection framework integrating a hybrid attention mechanism that synergistically combines channel-wise correlations and spatial dependencies. This dual attention design enables more discriminative feature extraction, which is particularly effective for distinguishing morphologically similar pest species. Third, we implement an optimized training strategy featuring a cosine annealing scheduler with linear warm-up, accelerating model convergence while maintaining training stability. Experiments have shown that compared with the original Faster RCNN model, GC-Faster RCNN has improved the average accuracy mAP0.5 on the Insect25 dataset by 4.5 percentage points, and mAP0.75 by 20.4 percentage points, mAP0.5:0.95 increased by 20.8 percentage points, and the recall rate increased by 16.6 percentage points. In addition, experiments have also shown that the GC-Faster RCNN detection method can reduce interference from multiple scales and high similarity between categories, improving detection performance. |
| format | Article |
| id | doaj-art-ec18dd1b60ee425889d9d690482dbe8e |
| institution | OA Journals |
| issn | 2223-7747 |
| language | English |
| publishDate | 2025-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Plants |
| spelling | doaj-art-ec18dd1b60ee425889d9d690482dbe8e2025-08-20T02:09:11ZengMDPI AGPlants2223-77472025-04-01147110610.3390/plants14071106GC-Faster RCNN: The Object Detection Algorithm for Agricultural Pests Based on Improved Hybrid Attention MechanismBolun Guan0Yaqian Wu1Jingbo Zhu2Juanjuan Kong3Wei Dong4Institute of Agricultural Economics and Information, Anhui Academy of Agricultural Sciences, Hefei 230031, ChinaSchool of Computer Science and Technology, Anhui University of Technology, Maanshan 243032, ChinaInstitute of Agricultural Economics and Information, Anhui Academy of Agricultural Sciences, Hefei 230031, ChinaInstitute of Agricultural Economics and Information, Anhui Academy of Agricultural Sciences, Hefei 230031, ChinaInstitute of Agricultural Economics and Information, Anhui Academy of Agricultural Sciences, Hefei 230031, ChinaPest infestations remain a critical threat to global agriculture, significantly compromising crop yield and quality. While accurate pest detection forms the foundation of precision pest management, current approaches face two primary challenges: (1) the scarcity of comprehensive multi-scale, multi-category pest datasets and (2) performance limitations in detection models caused by substantial target scale variations and high inter-class morphological similarity. To address these issues, we present three key contributions: First, we introduce Insect25—a novel agricultural pest detection dataset containing 25 distinct pest categories, comprising 18,349 high-resolution images. This dataset specifically addresses scale diversity through multi-resolution acquisition protocols, significantly enriching feature distribution for robust model training. Second, we propose GC-Faster RCNN, an enhanced detection framework integrating a hybrid attention mechanism that synergistically combines channel-wise correlations and spatial dependencies. This dual attention design enables more discriminative feature extraction, which is particularly effective for distinguishing morphologically similar pest species. Third, we implement an optimized training strategy featuring a cosine annealing scheduler with linear warm-up, accelerating model convergence while maintaining training stability. Experiments have shown that compared with the original Faster RCNN model, GC-Faster RCNN has improved the average accuracy mAP0.5 on the Insect25 dataset by 4.5 percentage points, and mAP0.75 by 20.4 percentage points, mAP0.5:0.95 increased by 20.8 percentage points, and the recall rate increased by 16.6 percentage points. In addition, experiments have also shown that the GC-Faster RCNN detection method can reduce interference from multiple scales and high similarity between categories, improving detection performance.https://www.mdpi.com/2223-7747/14/7/1106object detectionhybrid attention mechanismoptimization functionagricultural pests |
| spellingShingle | Bolun Guan Yaqian Wu Jingbo Zhu Juanjuan Kong Wei Dong GC-Faster RCNN: The Object Detection Algorithm for Agricultural Pests Based on Improved Hybrid Attention Mechanism Plants object detection hybrid attention mechanism optimization function agricultural pests |
| title | GC-Faster RCNN: The Object Detection Algorithm for Agricultural Pests Based on Improved Hybrid Attention Mechanism |
| title_full | GC-Faster RCNN: The Object Detection Algorithm for Agricultural Pests Based on Improved Hybrid Attention Mechanism |
| title_fullStr | GC-Faster RCNN: The Object Detection Algorithm for Agricultural Pests Based on Improved Hybrid Attention Mechanism |
| title_full_unstemmed | GC-Faster RCNN: The Object Detection Algorithm for Agricultural Pests Based on Improved Hybrid Attention Mechanism |
| title_short | GC-Faster RCNN: The Object Detection Algorithm for Agricultural Pests Based on Improved Hybrid Attention Mechanism |
| title_sort | gc faster rcnn the object detection algorithm for agricultural pests based on improved hybrid attention mechanism |
| topic | object detection hybrid attention mechanism optimization function agricultural pests |
| url | https://www.mdpi.com/2223-7747/14/7/1106 |
| work_keys_str_mv | AT bolunguan gcfasterrcnntheobjectdetectionalgorithmforagriculturalpestsbasedonimprovedhybridattentionmechanism AT yaqianwu gcfasterrcnntheobjectdetectionalgorithmforagriculturalpestsbasedonimprovedhybridattentionmechanism AT jingbozhu gcfasterrcnntheobjectdetectionalgorithmforagriculturalpestsbasedonimprovedhybridattentionmechanism AT juanjuankong gcfasterrcnntheobjectdetectionalgorithmforagriculturalpestsbasedonimprovedhybridattentionmechanism AT weidong gcfasterrcnntheobjectdetectionalgorithmforagriculturalpestsbasedonimprovedhybridattentionmechanism |